Maintained by Difan Deng and Marius Lindauer.
The following list considers papers related to neural architecture search. It is by no means complete. If you miss a paper on the list, please let us know.
Please note that although NAS methods steadily improve, the quality of empirical evaluations in this field are still lagging behind compared to other areas in machine learning, AI and optimization. We would therefore like to share some best practices for empirical evaluations of NAS methods, which we believe will facilitate sustained and measurable progress in the field. If you are interested in a teaser, please read our blog post or directly jump to our checklist.
Transformers have gained increasing popularity in different domains. For a comprehensive list of papers focusing on Neural Architecture Search for Transformer-Based spaces, the awesome-transformer-search repo is all you need.
1.
Berezsky, O. M.; Liashchynskyi, P. B.
METHOD OF GENERATIVE-ADVERSARIAL NETWORKS SEARCHING ARCHITECTURES FOR BIOMEDICAL IMAGES SYNTHESIS Journal Article
In: Radio Electronics, Computer Science, Control, no. 1, pp. 104, 2024.
@article{Berezsky_Liashchynskyi_2024,
title = {METHOD OF GENERATIVE-ADVERSARIAL NETWORKS SEARCHING ARCHITECTURES FOR BIOMEDICAL IMAGES SYNTHESIS},
author = {O. M. Berezsky and P. B. Liashchynskyi},
url = {http://ric.zntu.edu.ua/article/view/300976},
doi = {10.15588/1607-3274-2024-1-10},
year = {2024},
date = {2024-04-01},
urldate = {2024-04-01},
journal = {Radio Electronics, Computer Science, Control},
number = {1},
pages = {104},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
2.
Berezsky, Oleh; Liashchynskyi, Petro; Pitsun, Oleh; Izonin, Ivan
Synthesis of Convolutional Neural Network architectures for biomedical image classification Journal Article
In: Biomedical Signal Processing and Control, vol. 95, pp. 106325, 2024, ISSN: 1746-8094.
@article{BEREZSKY2024106325,
title = {Synthesis of Convolutional Neural Network architectures for biomedical image classification},
author = {Oleh Berezsky and Petro Liashchynskyi and Oleh Pitsun and Ivan Izonin},
url = {https://www.sciencedirect.com/science/article/pii/S1746809424003835},
doi = {https://doi.org/10.1016/j.bspc.2024.106325},
issn = {1746-8094},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {Biomedical Signal Processing and Control},
volume = {95},
pages = {106325},
abstract = {Convolutional Neural Networks (CNNs) are frequently used for image classification. This is crucial for the biomedical image classification used for automatic diagnosis in oncology. Designing optimal convolutional neural network architectures is a routine procedure that requires expert knowledge of computer vision and biomedical image features. To address this issue, we developed an automatic method for finding optimal CNN architectures. Our two-step method includes a genetic algorithm-based micro- and macro-search. Micro-search aims to find the optimal cell architecture based on the number of nodes and a set of predefined operations between nodes. Macro-search identifies the optimal number of cells and the operations between them to obtain the final optimal architecture. We obtained several optimal CNN architectures using the developed method of automatic architecture search. We conducted several computer experiments using cytological image classification as an example. The studies’ findings demonstrated that cytological image classification accuracy is higher compared to the classification accuracy of known CNN architectures (VGG-16, AlexNet, LeNet-5, ResNet-18, ResNet-50, MobileNetV3). The method is efficient because the search time for optimal architectures is short. Additionally, the method of optimal architecture search can be used for the synthesis of architectures used for the classification of other classes of biomedical images.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Convolutional Neural Networks (CNNs) are frequently used for image classification. This is crucial for the biomedical image classification used for automatic diagnosis in oncology. Designing optimal convolutional neural network architectures is a routine procedure that requires expert knowledge of computer vision and biomedical image features. To address this issue, we developed an automatic method for finding optimal CNN architectures. Our two-step method includes a genetic algorithm-based micro- and macro-search. Micro-search aims to find the optimal cell architecture based on the number of nodes and a set of predefined operations between nodes. Macro-search identifies the optimal number of cells and the operations between them to obtain the final optimal architecture. We obtained several optimal CNN architectures using the developed method of automatic architecture search. We conducted several computer experiments using cytological image classification as an example. The studies’ findings demonstrated that cytological image classification accuracy is higher compared to the classification accuracy of known CNN architectures (VGG-16, AlexNet, LeNet-5, ResNet-18, ResNet-50, MobileNetV3). The method is efficient because the search time for optimal architectures is short. Additionally, the method of optimal architecture search can be used for the synthesis of architectures used for the classification of other classes of biomedical images.
3.
Ali, Muhammad Junaid; Moalic, Laurent; Essaid, Mokhtar; Idoumghar, Lhassane
Robust Neural Architecture Search Using Differential Evolution for Medical Images Proceedings Article
In: Smith, Stephen; Correia, João; Cintrano, Christian (Ed.): Applications of Evolutionary Computation, pp. 163–179, Springer Nature Switzerland, Cham, 2024, ISBN: 978-3-031-56855-8.
@inproceedings{10.1007/978-3-031-56855-8_10,
title = {Robust Neural Architecture Search Using Differential Evolution for Medical Images},
author = {Muhammad Junaid Ali and Laurent Moalic and Mokhtar Essaid and Lhassane Idoumghar},
editor = {Stephen Smith and João Correia and Christian Cintrano},
url = {https://link.springer.com/chapter/10.1007/978-3-031-56855-8_10},
isbn = {978-3-031-56855-8},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {Applications of Evolutionary Computation},
pages = {163–179},
publisher = {Springer Nature Switzerland},
address = {Cham},
abstract = {Recent studies have demonstrated that Convolutional Neural Network (CNN) architectures are sensitive to adversarial attacks with imperceptible permutations. Adversarial attacks on medical images may cause manipulated decisions and decrease the performance of the diagnosis system. The robustness of medical systems is crucial, as it assures an improved healthcare system and assists medical professionals in making decisions. Various studies have been proposed to secure medical systems against adversarial attacks, but they have used handcrafted architectures. This study proposes an evolutionary Neural Architecture Search (NAS) approach for searching robust architectures for medical image classification. The Differential Evolution (DE) algorithm is used as a search algorithm. Furthermore, we utilize an attention-based search space consisting of five different attention layers and sixteen convolution and pooling operations. Experiments on multiple MedMNIST datasets show that the proposed approach has achieved better results than deep learning architectures and a robust NAS approach.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Recent studies have demonstrated that Convolutional Neural Network (CNN) architectures are sensitive to adversarial attacks with imperceptible permutations. Adversarial attacks on medical images may cause manipulated decisions and decrease the performance of the diagnosis system. The robustness of medical systems is crucial, as it assures an improved healthcare system and assists medical professionals in making decisions. Various studies have been proposed to secure medical systems against adversarial attacks, but they have used handcrafted architectures. This study proposes an evolutionary Neural Architecture Search (NAS) approach for searching robust architectures for medical image classification. The Differential Evolution (DE) algorithm is used as a search algorithm. Furthermore, we utilize an attention-based search space consisting of five different attention layers and sixteen convolution and pooling operations. Experiments on multiple MedMNIST datasets show that the proposed approach has achieved better results than deep learning architectures and a robust NAS approach.
4.
Liu, Xin; Tian, Jie; Duan, Peiyong; Yu, Qian; Wang, Gaige; Wang, Yingjie
GrMoNAS: A granularity-based multi-objective NAS framework for efficient medical diagnosis Journal Article
In: Computers in Biology and Medicine, vol. 171, pp. 108118, 2024, ISSN: 0010-4825.
@article{LIU2024108118,
title = {GrMoNAS: A granularity-based multi-objective NAS framework for efficient medical diagnosis},
author = {Xin Liu and Jie Tian and Peiyong Duan and Qian Yu and Gaige Wang and Yingjie Wang},
url = {https://www.sciencedirect.com/science/article/pii/S0010482524002026},
doi = {https://doi.org/10.1016/j.compbiomed.2024.108118},
issn = {0010-4825},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {Computers in Biology and Medicine},
volume = {171},
pages = {108118},
abstract = {Neural Architecture Search (NAS) has been widely applied to automate medical image diagnostics. However, traditional NAS methods require significant computational resources and time for performance evaluation. To address this, we introduce the GrMoNAS framework, designed to balance diagnostic accuracy and efficiency using proxy datasets for granularity transformation and multi-objective optimization algorithms. The approach initiates with a coarse granularity phase, wherein diverse candidate neural architectures undergo evaluation utilizing a reduced proxy dataset. This initial phase facilitates the swift and effective identification of architectures exhibiting promise. Subsequently, in the fine granularity phase, a comprehensive validation and optimization process is undertaken for these identified architectures. Concurrently, employing multi-objective optimization and Pareto frontier sorting aims to enhance both accuracy and computational efficiency simultaneously. Importantly, the GrMoNAS framework is particularly suitable for hospitals with limited computational resources. We evaluated GrMoNAS in a range of medical scenarios, such as COVID-19, Skin cancer, Lung, Colon, and Acute Lymphoblastic Leukemia diseases, comparing it against traditional models like VGG16, VGG19, and recent NAS approaches including GA-CNN, EBNAS, NEXception, and CovNAS. The results show that GrMoNAS achieves comparable or superior diagnostic precision, significantly enhancing diagnostic efficiency. Moreover, GrMoNAS effectively avoids local optima, indicating its significant potential for precision medical diagnosis.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Neural Architecture Search (NAS) has been widely applied to automate medical image diagnostics. However, traditional NAS methods require significant computational resources and time for performance evaluation. To address this, we introduce the GrMoNAS framework, designed to balance diagnostic accuracy and efficiency using proxy datasets for granularity transformation and multi-objective optimization algorithms. The approach initiates with a coarse granularity phase, wherein diverse candidate neural architectures undergo evaluation utilizing a reduced proxy dataset. This initial phase facilitates the swift and effective identification of architectures exhibiting promise. Subsequently, in the fine granularity phase, a comprehensive validation and optimization process is undertaken for these identified architectures. Concurrently, employing multi-objective optimization and Pareto frontier sorting aims to enhance both accuracy and computational efficiency simultaneously. Importantly, the GrMoNAS framework is particularly suitable for hospitals with limited computational resources. We evaluated GrMoNAS in a range of medical scenarios, such as COVID-19, Skin cancer, Lung, Colon, and Acute Lymphoblastic Leukemia diseases, comparing it against traditional models like VGG16, VGG19, and recent NAS approaches including GA-CNN, EBNAS, NEXception, and CovNAS. The results show that GrMoNAS achieves comparable or superior diagnostic precision, significantly enhancing diagnostic efficiency. Moreover, GrMoNAS effectively avoids local optima, indicating its significant potential for precision medical diagnosis.
5.
Cui, Suhan; Wang, Jiaqi; Zhong, Yuan; Liu, Han; Wang, Ting; Ma, Fenglong
Automated Fusion of Multimodal Electronic Health Records for Better Medical Predictions Technical Report
2024.
@techreport{cui2024automated,
title = {Automated Fusion of Multimodal Electronic Health Records for Better Medical Predictions},
author = {Suhan Cui and Jiaqi Wang and Yuan Zhong and Han Liu and Ting Wang and Fenglong Ma},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
6.
Fuentes-Tomás, José Antonio; Acosta-Mesa, Héctor Gabriel; Mezura-Montes, Efrén; Jiménez, Rodolfo Hernandez
Neural Architecture Search for Placenta Segmentation in 2D Ultrasound Images Proceedings Article
In: Calvo, Hiram; Martínez-Villaseñor, Lourdes; Ponce, Hiram; Cabada, Ramón Zatarain; Rivera, Martín Montes; Mezura-Montes, Efrén (Ed.): Advances in Computational Intelligence. MICAI 2023 International Workshops, pp. 397–408, Springer Nature Switzerland, Cham, 2024, ISBN: 978-3-031-51940-6.
@inproceedings{10.1007/978-3-031-51940-6_30,
title = {Neural Architecture Search for Placenta Segmentation in 2D Ultrasound Images},
author = {José Antonio Fuentes-Tomás and Héctor Gabriel Acosta-Mesa and Efrén Mezura-Montes and Rodolfo Hernandez Jiménez},
editor = {Hiram Calvo and Lourdes Martínez-Villaseñor and Hiram Ponce and Ramón Zatarain Cabada and Martín Montes Rivera and Efrén Mezura-Montes},
url = {https://link.springer.com/chapter/10.1007/978-3-031-51940-6_30#citeas},
isbn = {978-3-031-51940-6},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {Advances in Computational Intelligence. MICAI 2023 International Workshops},
pages = {397–408},
publisher = {Springer Nature Switzerland},
address = {Cham},
abstract = {Monitoring the placenta during pregnancy can lead to early diagnosis of anomalies by observing their characteristics, such as size, shape, and location. Ultrasound is a popular medical imaging technique used in placenta monitoring, whose advantages include the non-invasive feature, price, and accessibility. However, images from this domain are characterized by their noise. A segmentation system is required to recognize placenta features. U-Net architecture is a convolutional neural network that has become popular in the literature for medical image segmentation tasks. However, this type is a general-purpose network that requires great expertise to design and may only be applicable in some domains. The evolutionary computation overcomes this limitation, leading to the automatic design of convolutional neural networks. This work proposes a U-Net-based neural architecture search algorithm to construct convolutional neural networks applied in the placenta segmentation on 2D ultrasound images. The results show that the proposed algorithm allows a decrease in the number of parameters of U-Net, ranging from 80 to 98%. Moreover, the segmentation performance achieves a competitive level compared to U-Net, with a difference of 0.012 units in the Dice index.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Monitoring the placenta during pregnancy can lead to early diagnosis of anomalies by observing their characteristics, such as size, shape, and location. Ultrasound is a popular medical imaging technique used in placenta monitoring, whose advantages include the non-invasive feature, price, and accessibility. However, images from this domain are characterized by their noise. A segmentation system is required to recognize placenta features. U-Net architecture is a convolutional neural network that has become popular in the literature for medical image segmentation tasks. However, this type is a general-purpose network that requires great expertise to design and may only be applicable in some domains. The evolutionary computation overcomes this limitation, leading to the automatic design of convolutional neural networks. This work proposes a U-Net-based neural architecture search algorithm to construct convolutional neural networks applied in the placenta segmentation on 2D ultrasound images. The results show that the proposed algorithm allows a decrease in the number of parameters of U-Net, ranging from 80 to 98%. Moreover, the segmentation performance achieves a competitive level compared to U-Net, with a difference of 0.012 units in the Dice index.
7.
Liu, Jianwei Zhao Jie Li Xin
Evolutionary Neural Architecture Search and Its Applications in Healthcare Journal Article
In: Computer Modeling in Engineering & Sciences, vol. 139, no. 1, pp. 143–185, 2024, ISSN: 1526-1506.
@article{cmes.2023.030391,
title = {Evolutionary Neural Architecture Search and Its Applications in Healthcare},
author = {Jianwei Zhao Jie Li Xin Liu},
url = {http://www.techscience.com/CMES/v139n1/55101},
doi = {10.32604/cmes.2023.030391},
issn = {1526-1506},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {Computer Modeling in Engineering & Sciences},
volume = {139},
number = {1},
pages = {143–185},
abstract = {Most of the neural network architectures are based on human experience, which requires a long and tedious trial-and-error process. Neural architecture search (NAS) attempts to detect effective architectures without human intervention. Evolutionary algorithms (EAs) for NAS can find better solutions than human-designed architectures by exploring a large search space for possible architectures. Using multiobjective EAs for NAS, optimal neural architectures that meet various performance criteria can be explored and discovered efficiently. Furthermore, hardware-accelerated NAS methods can improve the efficiency of the NAS. While existing reviews have mainly focused on different strategies to complete NAS, a few studies have explored the use of EAs for NAS. In this paper, we summarize and explore the use of EAs for NAS, as well as large-scale multiobjective optimization strategies and hardware-accelerated NAS methods. NAS performs well in healthcare applications, such as medical image analysis, classification of disease diagnosis, and health monitoring. EAs for NAS can automate the search process and optimize multiple objectives simultaneously in a given healthcare task. Deep neural network has been successfully used in healthcare, but it lacks interpretability. Medical data is highly sensitive, and privacy leaks are frequently reported in the healthcare industry. To solve these problems, in healthcare, we propose an interpretable neuroevolution framework based on federated learning to address search efficiency and privacy protection. Moreover, we also point out future research directions for evolutionary NAS. Overall, for researchers who want to use EAs to optimize NNs in healthcare, we analyze the advantages and disadvantages of doing so to provide detailed guidance, and propose an interpretable privacy-preserving framework for healthcare applications.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Most of the neural network architectures are based on human experience, which requires a long and tedious trial-and-error process. Neural architecture search (NAS) attempts to detect effective architectures without human intervention. Evolutionary algorithms (EAs) for NAS can find better solutions than human-designed architectures by exploring a large search space for possible architectures. Using multiobjective EAs for NAS, optimal neural architectures that meet various performance criteria can be explored and discovered efficiently. Furthermore, hardware-accelerated NAS methods can improve the efficiency of the NAS. While existing reviews have mainly focused on different strategies to complete NAS, a few studies have explored the use of EAs for NAS. In this paper, we summarize and explore the use of EAs for NAS, as well as large-scale multiobjective optimization strategies and hardware-accelerated NAS methods. NAS performs well in healthcare applications, such as medical image analysis, classification of disease diagnosis, and health monitoring. EAs for NAS can automate the search process and optimize multiple objectives simultaneously in a given healthcare task. Deep neural network has been successfully used in healthcare, but it lacks interpretability. Medical data is highly sensitive, and privacy leaks are frequently reported in the healthcare industry. To solve these problems, in healthcare, we propose an interpretable neuroevolution framework based on federated learning to address search efficiency and privacy protection. Moreover, we also point out future research directions for evolutionary NAS. Overall, for researchers who want to use EAs to optimize NNs in healthcare, we analyze the advantages and disadvantages of doing so to provide detailed guidance, and propose an interpretable privacy-preserving framework for healthcare applications.
8.
Yang, Dong; Roth, Holger R.; Wang, Xiaosong; Xu, Ziyue; Xu, Daguang
In: Zhou, S. Kevin; Greenspan, Hayit; Shen, Dinggang (Ed.): Deep Learning for Medical Image Analysis (Second Edition), pp. 281-298, Academic Press, 2024, ISBN: 978-0-323-85124-4.
@incollection{YANG2024281,
title = {Chapter 10 - Dynamic inference using neural architecture search in medical image segmentation: From a novel adaptation perspective},
author = {Dong Yang and Holger R. Roth and Xiaosong Wang and Ziyue Xu and Daguang Xu},
editor = {S. Kevin Zhou and Hayit Greenspan and Dinggang Shen},
url = {https://www.sciencedirect.com/science/article/pii/B9780323851244000210},
doi = {https://doi.org/10.1016/B978-0-32-385124-4.00021-0},
isbn = {978-0-323-85124-4},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {Deep Learning for Medical Image Analysis (Second Edition)},
pages = {281-298},
publisher = {Academic Press},
edition = {Second Edition},
series = {The MICCAI Society book Series},
abstract = {Data inconsistency in medical imaging acquisition has been existing for decades, which creates difficulties when researchers adopt learning-based processing methods to unknown data. This issue is mostly caused by medical image scanners from different vendors, inconsistent scanning protocols, anatomy discrepancy among populations, environmental artifacts or other related factors. For instance, large appearance variance may exist in 3D T2-weighted brain MRI from different institutions or hospitals, even scanned with the same scanning protocols. Meanwhile, the data inconsistency downgrades the performance of machine learning models for medical image processing, such as organ or tumor segmentation, when models face unknown data at inference with pre-trained models. To alleviate the potential side effects caused by the data inconsistency, we propose a novel approach to improve model generalizability and transferability for unknown data leveraging the concepts from neural architecture search. We build a general “super-net” enabling multiple candidate modules in parallel to represent multi-scale contextual features at different network levels, respectively. After the training of the super-net is accomplished, a unique and optimal architecture for each data point is determined with guidance of additional model constraints at inference. We also propose a novel path sampling strategy to enable “fair” model training. Our experiments show that the proposed approach has clear advantages over the conventional neural network deployment in terms of segmentation performance and generalization in the unknown images.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
Data inconsistency in medical imaging acquisition has been existing for decades, which creates difficulties when researchers adopt learning-based processing methods to unknown data. This issue is mostly caused by medical image scanners from different vendors, inconsistent scanning protocols, anatomy discrepancy among populations, environmental artifacts or other related factors. For instance, large appearance variance may exist in 3D T2-weighted brain MRI from different institutions or hospitals, even scanned with the same scanning protocols. Meanwhile, the data inconsistency downgrades the performance of machine learning models for medical image processing, such as organ or tumor segmentation, when models face unknown data at inference with pre-trained models. To alleviate the potential side effects caused by the data inconsistency, we propose a novel approach to improve model generalizability and transferability for unknown data leveraging the concepts from neural architecture search. We build a general “super-net” enabling multiple candidate modules in parallel to represent multi-scale contextual features at different network levels, respectively. After the training of the super-net is accomplished, a unique and optimal architecture for each data point is determined with guidance of additional model constraints at inference. We also propose a novel path sampling strategy to enable “fair” model training. Our experiments show that the proposed approach has clear advantages over the conventional neural network deployment in terms of segmentation performance and generalization in the unknown images.
9.
Fuentes-Tomás, José-Antonio; Mezura-Montes, Efrén; Acosta-Mesa, Héctor-Gabriel; Márquez-Grajales, Aldo
Tree-Based Codification in Neural Architecture Search for Medical Image Segmentation Journal Article
In: IEEE Transactions on Evolutionary Computation, pp. 1-1, 2024.
@article{10391062,
title = {Tree-Based Codification in Neural Architecture Search for Medical Image Segmentation},
author = {José-Antonio Fuentes-Tomás and Efrén Mezura-Montes and Héctor-Gabriel Acosta-Mesa and Aldo Márquez-Grajales},
url = {https://ieeexplore.ieee.org/abstract/document/10391062},
doi = {10.1109/TEVC.2024.3353182},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {IEEE Transactions on Evolutionary Computation},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
10.
Wang, Yan; Zhen, Liangli; Zhang, Jianwei; Li, Miqing; Zhang, Lei; Wang, Zizhou; Feng, Yangqin; Xue, Yu; Wang, Xiao; Chen, Zheng; Luo, Tao; Goh, Rich Siow Mong; Liu, Yong
MedNAS: Multi-Scale Training-Free Neural Architecture Search for Medical Image Analysis Journal Article
In: IEEE Transactions on Evolutionary Computation, pp. 1-1, 2024.
@article{10391077,
title = {MedNAS: Multi-Scale Training-Free Neural Architecture Search for Medical Image Analysis},
author = {Yan Wang and Liangli Zhen and Jianwei Zhang and Miqing Li and Lei Zhang and Zizhou Wang and Yangqin Feng and Yu Xue and Xiao Wang and Zheng Chen and Tao Luo and Rich Siow Mong Goh and Yong Liu},
url = {https://ieeexplore.ieee.org/abstract/document/10391077/authors#authors},
doi = {10.1109/TEVC.2024.3352641},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {IEEE Transactions on Evolutionary Computation},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
11.
Şahin, Emrullah; Özdemir, Durmuş; Temurtaş, Hasan
Multi-objective optimization of ViT architecture for efficient brain tumor classification Journal Article
In: Biomedical Signal Processing and Control, vol. 91, pp. 105938, 2024, ISSN: 1746-8094.
@article{SAHIN2024105938,
title = {Multi-objective optimization of ViT architecture for efficient brain tumor classification},
author = {Emrullah Şahin and Durmuş Özdemir and Hasan Temurtaş},
url = {https://www.sciencedirect.com/science/article/pii/S174680942301371X},
doi = {https://doi.org/10.1016/j.bspc.2023.105938},
issn = {1746-8094},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {Biomedical Signal Processing and Control},
volume = {91},
pages = {105938},
abstract = {This study presents an advanced approach to optimizing the Vision Transformer (ViT) network for brain tumor classification in 2D MRI images, utilizing Bayesian Multi-Objective (BMO) optimization techniques. Rather than merely addressing the limitations of the standard ViT model, our objective was to enhance its overall efficiency and effectiveness. The application of BMO enabled us to fine-tune the architectural parameters of the ViT network, resulting in a model that was not only twice as fast but also four times smaller in size compared to the original. In terms of performance, the optimized ViT model achieved notable improvements, with a 1.48 % increase in validation accuracy, a 3.23 % rise in the F1-score, and a 3.36 % improvement in precision. These substantial enhancements highlight the potential of integrating BMO with visual transformer-based models, suggesting a promising direction for future research in achieving high efficiency and accuracy in complex classification tasks.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This study presents an advanced approach to optimizing the Vision Transformer (ViT) network for brain tumor classification in 2D MRI images, utilizing Bayesian Multi-Objective (BMO) optimization techniques. Rather than merely addressing the limitations of the standard ViT model, our objective was to enhance its overall efficiency and effectiveness. The application of BMO enabled us to fine-tune the architectural parameters of the ViT network, resulting in a model that was not only twice as fast but also four times smaller in size compared to the original. In terms of performance, the optimized ViT model achieved notable improvements, with a 1.48 % increase in validation accuracy, a 3.23 % rise in the F1-score, and a 3.36 % improvement in precision. These substantial enhancements highlight the potential of integrating BMO with visual transformer-based models, suggesting a promising direction for future research in achieving high efficiency and accuracy in complex classification tasks.
12.
Gao, Jianliang; Wu, Zhenpeng; Al-Sabri, Raeed; Oloulade, Babatounde Moctard; Chen, Jiamin
AutoDDI: Drug–Drug Interaction Prediction With Automated Graph Neural Network Journal Article
In: IEEE Journal of Biomedical and Health Informatics, pp. 1-12, 2024.
@article{10380606,
title = {AutoDDI: Drug–Drug Interaction Prediction With Automated Graph Neural Network},
author = {Jianliang Gao and Zhenpeng Wu and Raeed Al-Sabri and Babatounde Moctard Oloulade and Jiamin Chen},
doi = {10.1109/JBHI.2024.3349570},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {IEEE Journal of Biomedical and Health Informatics},
pages = {1-12},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
13.
He, Xin; Chu, Xiaowen
MedPipe: End-to-End Joint Search of Data Augmentation and Neural Architecture for 3D Medical Image Classification Journal Article
In: 2023.
@article{He_2023,
title = {MedPipe: End-to-End Joint Search of Data Augmentation and Neural Architecture for 3D Medical Image Classification},
author = {Xin He and Xiaowen Chu},
url = {http://dx.doi.org/10.36227/techrxiv.19513780.v2},
doi = {10.36227/techrxiv.19513780.v2},
year = {2023},
date = {2023-11-01},
urldate = {2023-11-01},
publisher = {Institute of Electrical and Electronics Engineers (IEEE)},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
14.
Akinola, Solomon Oluwole; Qingguo, Wang; Olukanmi, Peter; Tshilidzi, Marwala
A Boosted Evolutionary Neural Architecture Search for Timeseries Forecasting with Application to South African COVID-19 Cases Journal Article
In: International Journal of Online and Biomedical Engineering (iJOE), vol. 19, no. 14, pp. pp. 107–130, 2023.
@article{Akinola_Qingguo_Olukanmi_Tshilidzi_2023,
title = {A Boosted Evolutionary Neural Architecture Search for Timeseries Forecasting with Application to South African COVID-19 Cases},
author = {Solomon Oluwole Akinola and Wang Qingguo and Peter Olukanmi and Marwala Tshilidzi},
url = {https://online-journals.org/index.php/i-joe/article/view/41291},
doi = {10.3991/ijoe.v19i14.41291},
year = {2023},
date = {2023-10-01},
urldate = {2023-10-01},
journal = {International Journal of Online and Biomedical Engineering (iJOE)},
volume = {19},
number = {14},
pages = {pp. 107–130},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
15.
Zhang, Tian; Li, Nan; Zhou, Yuee; Cai, Wei; Ma, Lianbo
Information extraction of Chinese medical electronic records via evolutionary neural architecture search Proceedings Article
In: 2023 IEEE International Conference on Data Mining Workshops (ICDMW), pp. 396-405, 2023.
@inproceedings{10411614,
title = {Information extraction of Chinese medical electronic records via evolutionary neural architecture search},
author = {Tian Zhang and Nan Li and Yuee Zhou and Wei Cai and Lianbo Ma},
url = {https://ieeexplore.ieee.org/abstract/document/10411614},
doi = {10.1109/ICDMW60847.2023.00056},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE International Conference on Data Mining Workshops (ICDMW)},
pages = {396-405},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
16.
Meng, Lingtong; Chen, Yuting
DFairNAS: A Dataflow Fairness Approach to Training NAS Neural Networks Proceedings Article
In: 2023 16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), pp. 1-6, 2023.
@inproceedings{10373372,
title = {DFairNAS: A Dataflow Fairness Approach to Training NAS Neural Networks},
author = {Lingtong Meng and Yuting Chen},
url = {https://ieeexplore.ieee.org/abstract/document/10373372},
doi = {10.1109/CISP-BMEI60920.2023.10373372},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)},
pages = {1-6},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
17.
Mu, Pan; Wu, Guanyao; Liu, Jinyuan; Zhang, Yuduo; Fan, Xin; Liu, Risheng
Learning to Search a Lightweight Generalized Network for Medical Image Fusion Journal Article
In: IEEE Transactions on Circuits and Systems for Video Technology, pp. 1-1, 2023.
@article{10360160,
title = {Learning to Search a Lightweight Generalized Network for Medical Image Fusion},
author = {Pan Mu and Guanyao Wu and Jinyuan Liu and Yuduo Zhang and Xin Fan and Risheng Liu},
url = {https://ieeexplore.ieee.org/abstract/document/10360160},
doi = {10.1109/TCSVT.2023.3342808},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
18.
Li, J.; Lv, Y.; Zhang, P.; Zhao, J.
Neural Architecture Search for Unsupervised PET Image Denoising Proceedings Article
In: 2023 IEEE Nuclear Science Symposium, Medical Imaging Conference and International Symposium on Room-Temperature Semiconductor Detectors (NSS MIC RTSD), pp. 1-1, 2023.
@inproceedings{10338097,
title = {Neural Architecture Search for Unsupervised PET Image Denoising},
author = {J. Li and Y. Lv and P. Zhang and J. Zhao},
doi = {10.1109/NSSMICRTSD49126.2023.10338097},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE Nuclear Science Symposium, Medical Imaging Conference and International Symposium on Room-Temperature Semiconductor Detectors (NSS MIC RTSD)},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
19.
Yang, Changdi; Sheng, Yi; Dong, Peiyan; Kong, Zhenglun; Li, Yanyu; Yu, Pinrui; Yang, Lei; Lin, Xue; Wang, Yanzhi
Fast and Fair Medical AI on the Edge Through Neural Architecture Search for Hybrid Vision Models Proceedings Article
In: 2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD), pp. 01-09, 2023.
@inproceedings{10323652,
title = {Fast and Fair Medical AI on the Edge Through Neural Architecture Search for Hybrid Vision Models},
author = {Changdi Yang and Yi Sheng and Peiyan Dong and Zhenglun Kong and Yanyu Li and Pinrui Yu and Lei Yang and Xue Lin and Yanzhi Wang},
doi = {10.1109/ICCAD57390.2023.10323652},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD)},
pages = {01-09},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
20.
Liu, Xiao; Yao, Chong; Chen, Hongyi; Xiang, Rui; Wu, Hao; Du, Peng; Yu, Zekuan; Liu, Weifan; Liu, Jie; Geng, Daoying
BTSC-TNAS: A neural architecture search-based transformer for brain tumor segmentation and classification Journal Article
In: Computerized Medical Imaging and Graphics, vol. 110, pp. 102307, 2023, ISSN: 0895-6111.
@article{LIU2023102307,
title = {BTSC-TNAS: A neural architecture search-based transformer for brain tumor segmentation and classification},
author = {Xiao Liu and Chong Yao and Hongyi Chen and Rui Xiang and Hao Wu and Peng Du and Zekuan Yu and Weifan Liu and Jie Liu and Daoying Geng},
url = {https://www.sciencedirect.com/science/article/pii/S0895611123001258},
doi = {https://doi.org/10.1016/j.compmedimag.2023.102307},
issn = {0895-6111},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Computerized Medical Imaging and Graphics},
volume = {110},
pages = {102307},
abstract = {Glioblastoma (GBM), isolated brain metastasis (SBM), and primary central nervous system lymphoma (PCNSL) possess a high level of similarity in histomorphology and clinical manifestations on multimodal MRI. Such similarities have led to challenges in the clinical diagnosis of these three malignant tumors. However, many existing models solely focus on either the task of segmentation or classification, which limits the application of computer-aided diagnosis in clinical diagnosis and treatment. To solve this problem, we propose a multi-task learning transformer with neural architecture search (NAS) for brain tumor segmentation and classification (BTSC-TNAS). In the segmentation stage, we use a nested transformer U-shape network (NTU-NAS) with NAS to directly predict brain tumor masks from multi-modal MRI images. In the tumor classification stage, we use the multiscale features obtained from the encoder of NTU-NAS as the input features of the classification network (MSC-NET), which are integrated and corrected by the classification feature correction enhancement (CFCE) block to improve the accuracy of classification. The proposed BTSC-TNAS achieves an average Dice coefficient of 80.86% and 87.12% for the segmentation of tumor region and the maximum abnormal region in clinical data respectively. The model achieves a classification accuracy of 0.941. The experiments performed on the BraTS 2019 dataset show that the proposed BTSC-TNAS has excellent generalizability and can provide support for some challenging tasks in the diagnosis and treatment of brain tumors.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Glioblastoma (GBM), isolated brain metastasis (SBM), and primary central nervous system lymphoma (PCNSL) possess a high level of similarity in histomorphology and clinical manifestations on multimodal MRI. Such similarities have led to challenges in the clinical diagnosis of these three malignant tumors. However, many existing models solely focus on either the task of segmentation or classification, which limits the application of computer-aided diagnosis in clinical diagnosis and treatment. To solve this problem, we propose a multi-task learning transformer with neural architecture search (NAS) for brain tumor segmentation and classification (BTSC-TNAS). In the segmentation stage, we use a nested transformer U-shape network (NTU-NAS) with NAS to directly predict brain tumor masks from multi-modal MRI images. In the tumor classification stage, we use the multiscale features obtained from the encoder of NTU-NAS as the input features of the classification network (MSC-NET), which are integrated and corrected by the classification feature correction enhancement (CFCE) block to improve the accuracy of classification. The proposed BTSC-TNAS achieves an average Dice coefficient of 80.86% and 87.12% for the segmentation of tumor region and the maximum abnormal region in clinical data respectively. The model achieves a classification accuracy of 0.941. The experiments performed on the BraTS 2019 dataset show that the proposed BTSC-TNAS has excellent generalizability and can provide support for some challenging tasks in the diagnosis and treatment of brain tumors.
2024
113.
Berezsky, O. M.; Liashchynskyi, P. B.
METHOD OF GENERATIVE-ADVERSARIAL NETWORKS SEARCHING ARCHITECTURES FOR BIOMEDICAL IMAGES SYNTHESIS Journal Article
In: Radio Electronics, Computer Science, Control, no. 1, pp. 104, 2024.
@article{Berezsky_Liashchynskyi_2024,
title = {METHOD OF GENERATIVE-ADVERSARIAL NETWORKS SEARCHING ARCHITECTURES FOR BIOMEDICAL IMAGES SYNTHESIS},
author = {O. M. Berezsky and P. B. Liashchynskyi},
url = {http://ric.zntu.edu.ua/article/view/300976},
doi = {10.15588/1607-3274-2024-1-10},
year = {2024},
date = {2024-04-01},
urldate = {2024-04-01},
journal = {Radio Electronics, Computer Science, Control},
number = {1},
pages = {104},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
112.
Berezsky, Oleh; Liashchynskyi, Petro; Pitsun, Oleh; Izonin, Ivan
Synthesis of Convolutional Neural Network architectures for biomedical image classification Journal Article
In: Biomedical Signal Processing and Control, vol. 95, pp. 106325, 2024, ISSN: 1746-8094.
@article{BEREZSKY2024106325,
title = {Synthesis of Convolutional Neural Network architectures for biomedical image classification},
author = {Oleh Berezsky and Petro Liashchynskyi and Oleh Pitsun and Ivan Izonin},
url = {https://www.sciencedirect.com/science/article/pii/S1746809424003835},
doi = {https://doi.org/10.1016/j.bspc.2024.106325},
issn = {1746-8094},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {Biomedical Signal Processing and Control},
volume = {95},
pages = {106325},
abstract = {Convolutional Neural Networks (CNNs) are frequently used for image classification. This is crucial for the biomedical image classification used for automatic diagnosis in oncology. Designing optimal convolutional neural network architectures is a routine procedure that requires expert knowledge of computer vision and biomedical image features. To address this issue, we developed an automatic method for finding optimal CNN architectures. Our two-step method includes a genetic algorithm-based micro- and macro-search. Micro-search aims to find the optimal cell architecture based on the number of nodes and a set of predefined operations between nodes. Macro-search identifies the optimal number of cells and the operations between them to obtain the final optimal architecture. We obtained several optimal CNN architectures using the developed method of automatic architecture search. We conducted several computer experiments using cytological image classification as an example. The studies’ findings demonstrated that cytological image classification accuracy is higher compared to the classification accuracy of known CNN architectures (VGG-16, AlexNet, LeNet-5, ResNet-18, ResNet-50, MobileNetV3). The method is efficient because the search time for optimal architectures is short. Additionally, the method of optimal architecture search can be used for the synthesis of architectures used for the classification of other classes of biomedical images.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Convolutional Neural Networks (CNNs) are frequently used for image classification. This is crucial for the biomedical image classification used for automatic diagnosis in oncology. Designing optimal convolutional neural network architectures is a routine procedure that requires expert knowledge of computer vision and biomedical image features. To address this issue, we developed an automatic method for finding optimal CNN architectures. Our two-step method includes a genetic algorithm-based micro- and macro-search. Micro-search aims to find the optimal cell architecture based on the number of nodes and a set of predefined operations between nodes. Macro-search identifies the optimal number of cells and the operations between them to obtain the final optimal architecture. We obtained several optimal CNN architectures using the developed method of automatic architecture search. We conducted several computer experiments using cytological image classification as an example. The studies’ findings demonstrated that cytological image classification accuracy is higher compared to the classification accuracy of known CNN architectures (VGG-16, AlexNet, LeNet-5, ResNet-18, ResNet-50, MobileNetV3). The method is efficient because the search time for optimal architectures is short. Additionally, the method of optimal architecture search can be used for the synthesis of architectures used for the classification of other classes of biomedical images.
111.
Ali, Muhammad Junaid; Moalic, Laurent; Essaid, Mokhtar; Idoumghar, Lhassane
Robust Neural Architecture Search Using Differential Evolution for Medical Images Proceedings Article
In: Smith, Stephen; Correia, João; Cintrano, Christian (Ed.): Applications of Evolutionary Computation, pp. 163–179, Springer Nature Switzerland, Cham, 2024, ISBN: 978-3-031-56855-8.
@inproceedings{10.1007/978-3-031-56855-8_10,
title = {Robust Neural Architecture Search Using Differential Evolution for Medical Images},
author = {Muhammad Junaid Ali and Laurent Moalic and Mokhtar Essaid and Lhassane Idoumghar},
editor = {Stephen Smith and João Correia and Christian Cintrano},
url = {https://link.springer.com/chapter/10.1007/978-3-031-56855-8_10},
isbn = {978-3-031-56855-8},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {Applications of Evolutionary Computation},
pages = {163–179},
publisher = {Springer Nature Switzerland},
address = {Cham},
abstract = {Recent studies have demonstrated that Convolutional Neural Network (CNN) architectures are sensitive to adversarial attacks with imperceptible permutations. Adversarial attacks on medical images may cause manipulated decisions and decrease the performance of the diagnosis system. The robustness of medical systems is crucial, as it assures an improved healthcare system and assists medical professionals in making decisions. Various studies have been proposed to secure medical systems against adversarial attacks, but they have used handcrafted architectures. This study proposes an evolutionary Neural Architecture Search (NAS) approach for searching robust architectures for medical image classification. The Differential Evolution (DE) algorithm is used as a search algorithm. Furthermore, we utilize an attention-based search space consisting of five different attention layers and sixteen convolution and pooling operations. Experiments on multiple MedMNIST datasets show that the proposed approach has achieved better results than deep learning architectures and a robust NAS approach.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Recent studies have demonstrated that Convolutional Neural Network (CNN) architectures are sensitive to adversarial attacks with imperceptible permutations. Adversarial attacks on medical images may cause manipulated decisions and decrease the performance of the diagnosis system. The robustness of medical systems is crucial, as it assures an improved healthcare system and assists medical professionals in making decisions. Various studies have been proposed to secure medical systems against adversarial attacks, but they have used handcrafted architectures. This study proposes an evolutionary Neural Architecture Search (NAS) approach for searching robust architectures for medical image classification. The Differential Evolution (DE) algorithm is used as a search algorithm. Furthermore, we utilize an attention-based search space consisting of five different attention layers and sixteen convolution and pooling operations. Experiments on multiple MedMNIST datasets show that the proposed approach has achieved better results than deep learning architectures and a robust NAS approach.
110.
Liu, Xin; Tian, Jie; Duan, Peiyong; Yu, Qian; Wang, Gaige; Wang, Yingjie
GrMoNAS: A granularity-based multi-objective NAS framework for efficient medical diagnosis Journal Article
In: Computers in Biology and Medicine, vol. 171, pp. 108118, 2024, ISSN: 0010-4825.
@article{LIU2024108118,
title = {GrMoNAS: A granularity-based multi-objective NAS framework for efficient medical diagnosis},
author = {Xin Liu and Jie Tian and Peiyong Duan and Qian Yu and Gaige Wang and Yingjie Wang},
url = {https://www.sciencedirect.com/science/article/pii/S0010482524002026},
doi = {https://doi.org/10.1016/j.compbiomed.2024.108118},
issn = {0010-4825},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {Computers in Biology and Medicine},
volume = {171},
pages = {108118},
abstract = {Neural Architecture Search (NAS) has been widely applied to automate medical image diagnostics. However, traditional NAS methods require significant computational resources and time for performance evaluation. To address this, we introduce the GrMoNAS framework, designed to balance diagnostic accuracy and efficiency using proxy datasets for granularity transformation and multi-objective optimization algorithms. The approach initiates with a coarse granularity phase, wherein diverse candidate neural architectures undergo evaluation utilizing a reduced proxy dataset. This initial phase facilitates the swift and effective identification of architectures exhibiting promise. Subsequently, in the fine granularity phase, a comprehensive validation and optimization process is undertaken for these identified architectures. Concurrently, employing multi-objective optimization and Pareto frontier sorting aims to enhance both accuracy and computational efficiency simultaneously. Importantly, the GrMoNAS framework is particularly suitable for hospitals with limited computational resources. We evaluated GrMoNAS in a range of medical scenarios, such as COVID-19, Skin cancer, Lung, Colon, and Acute Lymphoblastic Leukemia diseases, comparing it against traditional models like VGG16, VGG19, and recent NAS approaches including GA-CNN, EBNAS, NEXception, and CovNAS. The results show that GrMoNAS achieves comparable or superior diagnostic precision, significantly enhancing diagnostic efficiency. Moreover, GrMoNAS effectively avoids local optima, indicating its significant potential for precision medical diagnosis.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Neural Architecture Search (NAS) has been widely applied to automate medical image diagnostics. However, traditional NAS methods require significant computational resources and time for performance evaluation. To address this, we introduce the GrMoNAS framework, designed to balance diagnostic accuracy and efficiency using proxy datasets for granularity transformation and multi-objective optimization algorithms. The approach initiates with a coarse granularity phase, wherein diverse candidate neural architectures undergo evaluation utilizing a reduced proxy dataset. This initial phase facilitates the swift and effective identification of architectures exhibiting promise. Subsequently, in the fine granularity phase, a comprehensive validation and optimization process is undertaken for these identified architectures. Concurrently, employing multi-objective optimization and Pareto frontier sorting aims to enhance both accuracy and computational efficiency simultaneously. Importantly, the GrMoNAS framework is particularly suitable for hospitals with limited computational resources. We evaluated GrMoNAS in a range of medical scenarios, such as COVID-19, Skin cancer, Lung, Colon, and Acute Lymphoblastic Leukemia diseases, comparing it against traditional models like VGG16, VGG19, and recent NAS approaches including GA-CNN, EBNAS, NEXception, and CovNAS. The results show that GrMoNAS achieves comparable or superior diagnostic precision, significantly enhancing diagnostic efficiency. Moreover, GrMoNAS effectively avoids local optima, indicating its significant potential for precision medical diagnosis.
109.
Cui, Suhan; Wang, Jiaqi; Zhong, Yuan; Liu, Han; Wang, Ting; Ma, Fenglong
Automated Fusion of Multimodal Electronic Health Records for Better Medical Predictions Technical Report
2024.
@techreport{cui2024automated,
title = {Automated Fusion of Multimodal Electronic Health Records for Better Medical Predictions},
author = {Suhan Cui and Jiaqi Wang and Yuan Zhong and Han Liu and Ting Wang and Fenglong Ma},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
108.
Fuentes-Tomás, José Antonio; Acosta-Mesa, Héctor Gabriel; Mezura-Montes, Efrén; Jiménez, Rodolfo Hernandez
Neural Architecture Search for Placenta Segmentation in 2D Ultrasound Images Proceedings Article
In: Calvo, Hiram; Martínez-Villaseñor, Lourdes; Ponce, Hiram; Cabada, Ramón Zatarain; Rivera, Martín Montes; Mezura-Montes, Efrén (Ed.): Advances in Computational Intelligence. MICAI 2023 International Workshops, pp. 397–408, Springer Nature Switzerland, Cham, 2024, ISBN: 978-3-031-51940-6.
@inproceedings{10.1007/978-3-031-51940-6_30,
title = {Neural Architecture Search for Placenta Segmentation in 2D Ultrasound Images},
author = {José Antonio Fuentes-Tomás and Héctor Gabriel Acosta-Mesa and Efrén Mezura-Montes and Rodolfo Hernandez Jiménez},
editor = {Hiram Calvo and Lourdes Martínez-Villaseñor and Hiram Ponce and Ramón Zatarain Cabada and Martín Montes Rivera and Efrén Mezura-Montes},
url = {https://link.springer.com/chapter/10.1007/978-3-031-51940-6_30#citeas},
isbn = {978-3-031-51940-6},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {Advances in Computational Intelligence. MICAI 2023 International Workshops},
pages = {397–408},
publisher = {Springer Nature Switzerland},
address = {Cham},
abstract = {Monitoring the placenta during pregnancy can lead to early diagnosis of anomalies by observing their characteristics, such as size, shape, and location. Ultrasound is a popular medical imaging technique used in placenta monitoring, whose advantages include the non-invasive feature, price, and accessibility. However, images from this domain are characterized by their noise. A segmentation system is required to recognize placenta features. U-Net architecture is a convolutional neural network that has become popular in the literature for medical image segmentation tasks. However, this type is a general-purpose network that requires great expertise to design and may only be applicable in some domains. The evolutionary computation overcomes this limitation, leading to the automatic design of convolutional neural networks. This work proposes a U-Net-based neural architecture search algorithm to construct convolutional neural networks applied in the placenta segmentation on 2D ultrasound images. The results show that the proposed algorithm allows a decrease in the number of parameters of U-Net, ranging from 80 to 98%. Moreover, the segmentation performance achieves a competitive level compared to U-Net, with a difference of 0.012 units in the Dice index.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Monitoring the placenta during pregnancy can lead to early diagnosis of anomalies by observing their characteristics, such as size, shape, and location. Ultrasound is a popular medical imaging technique used in placenta monitoring, whose advantages include the non-invasive feature, price, and accessibility. However, images from this domain are characterized by their noise. A segmentation system is required to recognize placenta features. U-Net architecture is a convolutional neural network that has become popular in the literature for medical image segmentation tasks. However, this type is a general-purpose network that requires great expertise to design and may only be applicable in some domains. The evolutionary computation overcomes this limitation, leading to the automatic design of convolutional neural networks. This work proposes a U-Net-based neural architecture search algorithm to construct convolutional neural networks applied in the placenta segmentation on 2D ultrasound images. The results show that the proposed algorithm allows a decrease in the number of parameters of U-Net, ranging from 80 to 98%. Moreover, the segmentation performance achieves a competitive level compared to U-Net, with a difference of 0.012 units in the Dice index.
107.
Liu, Jianwei Zhao Jie Li Xin
Evolutionary Neural Architecture Search and Its Applications in Healthcare Journal Article
In: Computer Modeling in Engineering & Sciences, vol. 139, no. 1, pp. 143–185, 2024, ISSN: 1526-1506.
@article{cmes.2023.030391,
title = {Evolutionary Neural Architecture Search and Its Applications in Healthcare},
author = {Jianwei Zhao Jie Li Xin Liu},
url = {http://www.techscience.com/CMES/v139n1/55101},
doi = {10.32604/cmes.2023.030391},
issn = {1526-1506},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {Computer Modeling in Engineering & Sciences},
volume = {139},
number = {1},
pages = {143–185},
abstract = {Most of the neural network architectures are based on human experience, which requires a long and tedious trial-and-error process. Neural architecture search (NAS) attempts to detect effective architectures without human intervention. Evolutionary algorithms (EAs) for NAS can find better solutions than human-designed architectures by exploring a large search space for possible architectures. Using multiobjective EAs for NAS, optimal neural architectures that meet various performance criteria can be explored and discovered efficiently. Furthermore, hardware-accelerated NAS methods can improve the efficiency of the NAS. While existing reviews have mainly focused on different strategies to complete NAS, a few studies have explored the use of EAs for NAS. In this paper, we summarize and explore the use of EAs for NAS, as well as large-scale multiobjective optimization strategies and hardware-accelerated NAS methods. NAS performs well in healthcare applications, such as medical image analysis, classification of disease diagnosis, and health monitoring. EAs for NAS can automate the search process and optimize multiple objectives simultaneously in a given healthcare task. Deep neural network has been successfully used in healthcare, but it lacks interpretability. Medical data is highly sensitive, and privacy leaks are frequently reported in the healthcare industry. To solve these problems, in healthcare, we propose an interpretable neuroevolution framework based on federated learning to address search efficiency and privacy protection. Moreover, we also point out future research directions for evolutionary NAS. Overall, for researchers who want to use EAs to optimize NNs in healthcare, we analyze the advantages and disadvantages of doing so to provide detailed guidance, and propose an interpretable privacy-preserving framework for healthcare applications.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Most of the neural network architectures are based on human experience, which requires a long and tedious trial-and-error process. Neural architecture search (NAS) attempts to detect effective architectures without human intervention. Evolutionary algorithms (EAs) for NAS can find better solutions than human-designed architectures by exploring a large search space for possible architectures. Using multiobjective EAs for NAS, optimal neural architectures that meet various performance criteria can be explored and discovered efficiently. Furthermore, hardware-accelerated NAS methods can improve the efficiency of the NAS. While existing reviews have mainly focused on different strategies to complete NAS, a few studies have explored the use of EAs for NAS. In this paper, we summarize and explore the use of EAs for NAS, as well as large-scale multiobjective optimization strategies and hardware-accelerated NAS methods. NAS performs well in healthcare applications, such as medical image analysis, classification of disease diagnosis, and health monitoring. EAs for NAS can automate the search process and optimize multiple objectives simultaneously in a given healthcare task. Deep neural network has been successfully used in healthcare, but it lacks interpretability. Medical data is highly sensitive, and privacy leaks are frequently reported in the healthcare industry. To solve these problems, in healthcare, we propose an interpretable neuroevolution framework based on federated learning to address search efficiency and privacy protection. Moreover, we also point out future research directions for evolutionary NAS. Overall, for researchers who want to use EAs to optimize NNs in healthcare, we analyze the advantages and disadvantages of doing so to provide detailed guidance, and propose an interpretable privacy-preserving framework for healthcare applications.
106.
Yang, Dong; Roth, Holger R.; Wang, Xiaosong; Xu, Ziyue; Xu, Daguang
In: Zhou, S. Kevin; Greenspan, Hayit; Shen, Dinggang (Ed.): Deep Learning for Medical Image Analysis (Second Edition), pp. 281-298, Academic Press, 2024, ISBN: 978-0-323-85124-4.
@incollection{YANG2024281,
title = {Chapter 10 - Dynamic inference using neural architecture search in medical image segmentation: From a novel adaptation perspective},
author = {Dong Yang and Holger R. Roth and Xiaosong Wang and Ziyue Xu and Daguang Xu},
editor = {S. Kevin Zhou and Hayit Greenspan and Dinggang Shen},
url = {https://www.sciencedirect.com/science/article/pii/B9780323851244000210},
doi = {https://doi.org/10.1016/B978-0-32-385124-4.00021-0},
isbn = {978-0-323-85124-4},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {Deep Learning for Medical Image Analysis (Second Edition)},
pages = {281-298},
publisher = {Academic Press},
edition = {Second Edition},
series = {The MICCAI Society book Series},
abstract = {Data inconsistency in medical imaging acquisition has been existing for decades, which creates difficulties when researchers adopt learning-based processing methods to unknown data. This issue is mostly caused by medical image scanners from different vendors, inconsistent scanning protocols, anatomy discrepancy among populations, environmental artifacts or other related factors. For instance, large appearance variance may exist in 3D T2-weighted brain MRI from different institutions or hospitals, even scanned with the same scanning protocols. Meanwhile, the data inconsistency downgrades the performance of machine learning models for medical image processing, such as organ or tumor segmentation, when models face unknown data at inference with pre-trained models. To alleviate the potential side effects caused by the data inconsistency, we propose a novel approach to improve model generalizability and transferability for unknown data leveraging the concepts from neural architecture search. We build a general “super-net” enabling multiple candidate modules in parallel to represent multi-scale contextual features at different network levels, respectively. After the training of the super-net is accomplished, a unique and optimal architecture for each data point is determined with guidance of additional model constraints at inference. We also propose a novel path sampling strategy to enable “fair” model training. Our experiments show that the proposed approach has clear advantages over the conventional neural network deployment in terms of segmentation performance and generalization in the unknown images.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
Data inconsistency in medical imaging acquisition has been existing for decades, which creates difficulties when researchers adopt learning-based processing methods to unknown data. This issue is mostly caused by medical image scanners from different vendors, inconsistent scanning protocols, anatomy discrepancy among populations, environmental artifacts or other related factors. For instance, large appearance variance may exist in 3D T2-weighted brain MRI from different institutions or hospitals, even scanned with the same scanning protocols. Meanwhile, the data inconsistency downgrades the performance of machine learning models for medical image processing, such as organ or tumor segmentation, when models face unknown data at inference with pre-trained models. To alleviate the potential side effects caused by the data inconsistency, we propose a novel approach to improve model generalizability and transferability for unknown data leveraging the concepts from neural architecture search. We build a general “super-net” enabling multiple candidate modules in parallel to represent multi-scale contextual features at different network levels, respectively. After the training of the super-net is accomplished, a unique and optimal architecture for each data point is determined with guidance of additional model constraints at inference. We also propose a novel path sampling strategy to enable “fair” model training. Our experiments show that the proposed approach has clear advantages over the conventional neural network deployment in terms of segmentation performance and generalization in the unknown images.
105.
Fuentes-Tomás, José-Antonio; Mezura-Montes, Efrén; Acosta-Mesa, Héctor-Gabriel; Márquez-Grajales, Aldo
Tree-Based Codification in Neural Architecture Search for Medical Image Segmentation Journal Article
In: IEEE Transactions on Evolutionary Computation, pp. 1-1, 2024.
@article{10391062,
title = {Tree-Based Codification in Neural Architecture Search for Medical Image Segmentation},
author = {José-Antonio Fuentes-Tomás and Efrén Mezura-Montes and Héctor-Gabriel Acosta-Mesa and Aldo Márquez-Grajales},
url = {https://ieeexplore.ieee.org/abstract/document/10391062},
doi = {10.1109/TEVC.2024.3353182},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {IEEE Transactions on Evolutionary Computation},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
104.
Wang, Yan; Zhen, Liangli; Zhang, Jianwei; Li, Miqing; Zhang, Lei; Wang, Zizhou; Feng, Yangqin; Xue, Yu; Wang, Xiao; Chen, Zheng; Luo, Tao; Goh, Rich Siow Mong; Liu, Yong
MedNAS: Multi-Scale Training-Free Neural Architecture Search for Medical Image Analysis Journal Article
In: IEEE Transactions on Evolutionary Computation, pp. 1-1, 2024.
@article{10391077,
title = {MedNAS: Multi-Scale Training-Free Neural Architecture Search for Medical Image Analysis},
author = {Yan Wang and Liangli Zhen and Jianwei Zhang and Miqing Li and Lei Zhang and Zizhou Wang and Yangqin Feng and Yu Xue and Xiao Wang and Zheng Chen and Tao Luo and Rich Siow Mong Goh and Yong Liu},
url = {https://ieeexplore.ieee.org/abstract/document/10391077/authors#authors},
doi = {10.1109/TEVC.2024.3352641},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {IEEE Transactions on Evolutionary Computation},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
103.
Şahin, Emrullah; Özdemir, Durmuş; Temurtaş, Hasan
Multi-objective optimization of ViT architecture for efficient brain tumor classification Journal Article
In: Biomedical Signal Processing and Control, vol. 91, pp. 105938, 2024, ISSN: 1746-8094.
@article{SAHIN2024105938,
title = {Multi-objective optimization of ViT architecture for efficient brain tumor classification},
author = {Emrullah Şahin and Durmuş Özdemir and Hasan Temurtaş},
url = {https://www.sciencedirect.com/science/article/pii/S174680942301371X},
doi = {https://doi.org/10.1016/j.bspc.2023.105938},
issn = {1746-8094},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {Biomedical Signal Processing and Control},
volume = {91},
pages = {105938},
abstract = {This study presents an advanced approach to optimizing the Vision Transformer (ViT) network for brain tumor classification in 2D MRI images, utilizing Bayesian Multi-Objective (BMO) optimization techniques. Rather than merely addressing the limitations of the standard ViT model, our objective was to enhance its overall efficiency and effectiveness. The application of BMO enabled us to fine-tune the architectural parameters of the ViT network, resulting in a model that was not only twice as fast but also four times smaller in size compared to the original. In terms of performance, the optimized ViT model achieved notable improvements, with a 1.48 % increase in validation accuracy, a 3.23 % rise in the F1-score, and a 3.36 % improvement in precision. These substantial enhancements highlight the potential of integrating BMO with visual transformer-based models, suggesting a promising direction for future research in achieving high efficiency and accuracy in complex classification tasks.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This study presents an advanced approach to optimizing the Vision Transformer (ViT) network for brain tumor classification in 2D MRI images, utilizing Bayesian Multi-Objective (BMO) optimization techniques. Rather than merely addressing the limitations of the standard ViT model, our objective was to enhance its overall efficiency and effectiveness. The application of BMO enabled us to fine-tune the architectural parameters of the ViT network, resulting in a model that was not only twice as fast but also four times smaller in size compared to the original. In terms of performance, the optimized ViT model achieved notable improvements, with a 1.48 % increase in validation accuracy, a 3.23 % rise in the F1-score, and a 3.36 % improvement in precision. These substantial enhancements highlight the potential of integrating BMO with visual transformer-based models, suggesting a promising direction for future research in achieving high efficiency and accuracy in complex classification tasks.
102.
Gao, Jianliang; Wu, Zhenpeng; Al-Sabri, Raeed; Oloulade, Babatounde Moctard; Chen, Jiamin
AutoDDI: Drug–Drug Interaction Prediction With Automated Graph Neural Network Journal Article
In: IEEE Journal of Biomedical and Health Informatics, pp. 1-12, 2024.
@article{10380606,
title = {AutoDDI: Drug–Drug Interaction Prediction With Automated Graph Neural Network},
author = {Jianliang Gao and Zhenpeng Wu and Raeed Al-Sabri and Babatounde Moctard Oloulade and Jiamin Chen},
doi = {10.1109/JBHI.2024.3349570},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {IEEE Journal of Biomedical and Health Informatics},
pages = {1-12},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
2023
101.
He, Xin; Chu, Xiaowen
MedPipe: End-to-End Joint Search of Data Augmentation and Neural Architecture for 3D Medical Image Classification Journal Article
In: 2023.
@article{He_2023,
title = {MedPipe: End-to-End Joint Search of Data Augmentation and Neural Architecture for 3D Medical Image Classification},
author = {Xin He and Xiaowen Chu},
url = {http://dx.doi.org/10.36227/techrxiv.19513780.v2},
doi = {10.36227/techrxiv.19513780.v2},
year = {2023},
date = {2023-11-01},
urldate = {2023-11-01},
publisher = {Institute of Electrical and Electronics Engineers (IEEE)},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
100.
Akinola, Solomon Oluwole; Qingguo, Wang; Olukanmi, Peter; Tshilidzi, Marwala
A Boosted Evolutionary Neural Architecture Search for Timeseries Forecasting with Application to South African COVID-19 Cases Journal Article
In: International Journal of Online and Biomedical Engineering (iJOE), vol. 19, no. 14, pp. pp. 107–130, 2023.
@article{Akinola_Qingguo_Olukanmi_Tshilidzi_2023,
title = {A Boosted Evolutionary Neural Architecture Search for Timeseries Forecasting with Application to South African COVID-19 Cases},
author = {Solomon Oluwole Akinola and Wang Qingguo and Peter Olukanmi and Marwala Tshilidzi},
url = {https://online-journals.org/index.php/i-joe/article/view/41291},
doi = {10.3991/ijoe.v19i14.41291},
year = {2023},
date = {2023-10-01},
urldate = {2023-10-01},
journal = {International Journal of Online and Biomedical Engineering (iJOE)},
volume = {19},
number = {14},
pages = {pp. 107–130},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
99.
Zhang, Tian; Li, Nan; Zhou, Yuee; Cai, Wei; Ma, Lianbo
Information extraction of Chinese medical electronic records via evolutionary neural architecture search Proceedings Article
In: 2023 IEEE International Conference on Data Mining Workshops (ICDMW), pp. 396-405, 2023.
@inproceedings{10411614,
title = {Information extraction of Chinese medical electronic records via evolutionary neural architecture search},
author = {Tian Zhang and Nan Li and Yuee Zhou and Wei Cai and Lianbo Ma},
url = {https://ieeexplore.ieee.org/abstract/document/10411614},
doi = {10.1109/ICDMW60847.2023.00056},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE International Conference on Data Mining Workshops (ICDMW)},
pages = {396-405},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
98.
Meng, Lingtong; Chen, Yuting
DFairNAS: A Dataflow Fairness Approach to Training NAS Neural Networks Proceedings Article
In: 2023 16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), pp. 1-6, 2023.
@inproceedings{10373372,
title = {DFairNAS: A Dataflow Fairness Approach to Training NAS Neural Networks},
author = {Lingtong Meng and Yuting Chen},
url = {https://ieeexplore.ieee.org/abstract/document/10373372},
doi = {10.1109/CISP-BMEI60920.2023.10373372},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)},
pages = {1-6},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
97.
Mu, Pan; Wu, Guanyao; Liu, Jinyuan; Zhang, Yuduo; Fan, Xin; Liu, Risheng
Learning to Search a Lightweight Generalized Network for Medical Image Fusion Journal Article
In: IEEE Transactions on Circuits and Systems for Video Technology, pp. 1-1, 2023.
@article{10360160,
title = {Learning to Search a Lightweight Generalized Network for Medical Image Fusion},
author = {Pan Mu and Guanyao Wu and Jinyuan Liu and Yuduo Zhang and Xin Fan and Risheng Liu},
url = {https://ieeexplore.ieee.org/abstract/document/10360160},
doi = {10.1109/TCSVT.2023.3342808},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
96.
Li, J.; Lv, Y.; Zhang, P.; Zhao, J.
Neural Architecture Search for Unsupervised PET Image Denoising Proceedings Article
In: 2023 IEEE Nuclear Science Symposium, Medical Imaging Conference and International Symposium on Room-Temperature Semiconductor Detectors (NSS MIC RTSD), pp. 1-1, 2023.
@inproceedings{10338097,
title = {Neural Architecture Search for Unsupervised PET Image Denoising},
author = {J. Li and Y. Lv and P. Zhang and J. Zhao},
doi = {10.1109/NSSMICRTSD49126.2023.10338097},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE Nuclear Science Symposium, Medical Imaging Conference and International Symposium on Room-Temperature Semiconductor Detectors (NSS MIC RTSD)},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
95.
Yang, Changdi; Sheng, Yi; Dong, Peiyan; Kong, Zhenglun; Li, Yanyu; Yu, Pinrui; Yang, Lei; Lin, Xue; Wang, Yanzhi
Fast and Fair Medical AI on the Edge Through Neural Architecture Search for Hybrid Vision Models Proceedings Article
In: 2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD), pp. 01-09, 2023.
@inproceedings{10323652,
title = {Fast and Fair Medical AI on the Edge Through Neural Architecture Search for Hybrid Vision Models},
author = {Changdi Yang and Yi Sheng and Peiyan Dong and Zhenglun Kong and Yanyu Li and Pinrui Yu and Lei Yang and Xue Lin and Yanzhi Wang},
doi = {10.1109/ICCAD57390.2023.10323652},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD)},
pages = {01-09},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
94.
Liu, Xiao; Yao, Chong; Chen, Hongyi; Xiang, Rui; Wu, Hao; Du, Peng; Yu, Zekuan; Liu, Weifan; Liu, Jie; Geng, Daoying
BTSC-TNAS: A neural architecture search-based transformer for brain tumor segmentation and classification Journal Article
In: Computerized Medical Imaging and Graphics, vol. 110, pp. 102307, 2023, ISSN: 0895-6111.
@article{LIU2023102307,
title = {BTSC-TNAS: A neural architecture search-based transformer for brain tumor segmentation and classification},
author = {Xiao Liu and Chong Yao and Hongyi Chen and Rui Xiang and Hao Wu and Peng Du and Zekuan Yu and Weifan Liu and Jie Liu and Daoying Geng},
url = {https://www.sciencedirect.com/science/article/pii/S0895611123001258},
doi = {https://doi.org/10.1016/j.compmedimag.2023.102307},
issn = {0895-6111},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Computerized Medical Imaging and Graphics},
volume = {110},
pages = {102307},
abstract = {Glioblastoma (GBM), isolated brain metastasis (SBM), and primary central nervous system lymphoma (PCNSL) possess a high level of similarity in histomorphology and clinical manifestations on multimodal MRI. Such similarities have led to challenges in the clinical diagnosis of these three malignant tumors. However, many existing models solely focus on either the task of segmentation or classification, which limits the application of computer-aided diagnosis in clinical diagnosis and treatment. To solve this problem, we propose a multi-task learning transformer with neural architecture search (NAS) for brain tumor segmentation and classification (BTSC-TNAS). In the segmentation stage, we use a nested transformer U-shape network (NTU-NAS) with NAS to directly predict brain tumor masks from multi-modal MRI images. In the tumor classification stage, we use the multiscale features obtained from the encoder of NTU-NAS as the input features of the classification network (MSC-NET), which are integrated and corrected by the classification feature correction enhancement (CFCE) block to improve the accuracy of classification. The proposed BTSC-TNAS achieves an average Dice coefficient of 80.86% and 87.12% for the segmentation of tumor region and the maximum abnormal region in clinical data respectively. The model achieves a classification accuracy of 0.941. The experiments performed on the BraTS 2019 dataset show that the proposed BTSC-TNAS has excellent generalizability and can provide support for some challenging tasks in the diagnosis and treatment of brain tumors.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Glioblastoma (GBM), isolated brain metastasis (SBM), and primary central nervous system lymphoma (PCNSL) possess a high level of similarity in histomorphology and clinical manifestations on multimodal MRI. Such similarities have led to challenges in the clinical diagnosis of these three malignant tumors. However, many existing models solely focus on either the task of segmentation or classification, which limits the application of computer-aided diagnosis in clinical diagnosis and treatment. To solve this problem, we propose a multi-task learning transformer with neural architecture search (NAS) for brain tumor segmentation and classification (BTSC-TNAS). In the segmentation stage, we use a nested transformer U-shape network (NTU-NAS) with NAS to directly predict brain tumor masks from multi-modal MRI images. In the tumor classification stage, we use the multiscale features obtained from the encoder of NTU-NAS as the input features of the classification network (MSC-NET), which are integrated and corrected by the classification feature correction enhancement (CFCE) block to improve the accuracy of classification. The proposed BTSC-TNAS achieves an average Dice coefficient of 80.86% and 87.12% for the segmentation of tumor region and the maximum abnormal region in clinical data respectively. The model achieves a classification accuracy of 0.941. The experiments performed on the BraTS 2019 dataset show that the proposed BTSC-TNAS has excellent generalizability and can provide support for some challenging tasks in the diagnosis and treatment of brain tumors.
93.
Yu, Caiyang; Wang, Yixi; Tang, Chenwei; Feng, Wentao; Lv, Jiancheng
EU-Net: Automatic U-Net neural architecture search with differential evolutionary algorithm for medical image segmentation Journal Article
In: Computers in Biology and Medicine, vol. 167, pp. 107579, 2023, ISSN: 0010-4825.
@article{YU2023107579,
title = {EU-Net: Automatic U-Net neural architecture search with differential evolutionary algorithm for medical image segmentation},
author = {Caiyang Yu and Yixi Wang and Chenwei Tang and Wentao Feng and Jiancheng Lv},
url = {https://www.sciencedirect.com/science/article/pii/S0010482523010442},
doi = {https://doi.org/10.1016/j.compbiomed.2023.107579},
issn = {0010-4825},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Computers in Biology and Medicine},
volume = {167},
pages = {107579},
abstract = {Medical images are crucial in clinical practice, providing essential information for patient assessment and treatment planning. However, manual extraction of information from images is both time-consuming and prone to errors. The emergence of U-Net addresses this challenge by automating the segmentation of anatomical structures and pathological lesions in medical images, thereby significantly enhancing the accuracy of image interpretation and diagnosis. However, the performance of U-Net largely depends on its encoder–decoder structure, which requires researchers with knowledge of neural network architecture design and an in-depth understanding of medical images. In this paper, we propose an automatic U-Net Neural Architecture Search (NAS) algorithm using the differential evolutionary (DE) algorithm, named EU-Net, to segment critical information in medical images to assist physicians in diagnosis. Specifically, by presenting the variable-length strategy, the proposed EU-Net algorithm can sufficiently and automatically search for the neural network architecture without expertise. Moreover, the utilization of crossover, mutation, and selection strategies of DE takes account of the trade-off between exploration and exploitation in the search space. Finally, in the encoding and decoding phases of the proposed algorithm, different block-based and layer-based structures are introduced for architectural optimization. The proposed EU-Net algorithm is validated on two widely used medical datasets, i.e., CHAOS and BUSI, for image segmentation tasks. Extensive experimental results show that the proposed EU-Net algorithm outperforms the chosen peer competitors in both two datasets. In particular, compared to the original U-Net, our proposed method improves the metric mIou by at least 6%.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Medical images are crucial in clinical practice, providing essential information for patient assessment and treatment planning. However, manual extraction of information from images is both time-consuming and prone to errors. The emergence of U-Net addresses this challenge by automating the segmentation of anatomical structures and pathological lesions in medical images, thereby significantly enhancing the accuracy of image interpretation and diagnosis. However, the performance of U-Net largely depends on its encoder–decoder structure, which requires researchers with knowledge of neural network architecture design and an in-depth understanding of medical images. In this paper, we propose an automatic U-Net Neural Architecture Search (NAS) algorithm using the differential evolutionary (DE) algorithm, named EU-Net, to segment critical information in medical images to assist physicians in diagnosis. Specifically, by presenting the variable-length strategy, the proposed EU-Net algorithm can sufficiently and automatically search for the neural network architecture without expertise. Moreover, the utilization of crossover, mutation, and selection strategies of DE takes account of the trade-off between exploration and exploitation in the search space. Finally, in the encoding and decoding phases of the proposed algorithm, different block-based and layer-based structures are introduced for architectural optimization. The proposed EU-Net algorithm is validated on two widely used medical datasets, i.e., CHAOS and BUSI, for image segmentation tasks. Extensive experimental results show that the proposed EU-Net algorithm outperforms the chosen peer competitors in both two datasets. In particular, compared to the original U-Net, our proposed method improves the metric mIou by at least 6%.
92.
Ghosh, Arjun; Jana, Nanda Dulal; Das, Swagatam; Mallipeddi, Rammohan
Two-Phase Evolutionary Convolutional Neural Network Architecture Search for Medical Image Classification Journal Article
In: IEEE Access, vol. 11, pp. 115280–115305, 2023.
@article{DBLP:journals/access/GhoshJDM23,
title = {Two-Phase Evolutionary Convolutional Neural Network Architecture Search for Medical Image Classification},
author = {Arjun Ghosh and Nanda Dulal Jana and Swagatam Das and Rammohan Mallipeddi},
url = {https://doi.org/10.1109/ACCESS.2023.3323705},
doi = {10.1109/ACCESS.2023.3323705},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Access},
volume = {11},
pages = {115280–115305},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
91.
Ghosh, Arjun; Jana, Nanda Dulal; Das, Swagatam; Mallipeddi, Rammohan
Two-Phase Evolutionary Convolutional Neural Network Architecture Search for Medical Image Classification Journal Article
In: IEEE Access, vol. 11, pp. 115280-115305, 2023.
@article{10278411,
title = {Two-Phase Evolutionary Convolutional Neural Network Architecture Search for Medical Image Classification},
author = {Arjun Ghosh and Nanda Dulal Jana and Swagatam Das and Rammohan Mallipeddi},
url = {https://ieeexplore.ieee.org/document/10278411},
doi = {10.1109/ACCESS.2023.3323705},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Access},
volume = {11},
pages = {115280-115305},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
90.
Xu, Hongyan; Wang, Dadong; Sowmya, Arcot; Katz, Ian
Detection of Basal Cell Carcinoma in Whole Slide Images Proceedings Article
In: Greenspan, Hayit; Madabhushi, Anant; Mousavi, Parvin; Salcudean, Septimiu; Duncan, James; Syeda-Mahmood, Tanveer; Taylor, Russell (Ed.): Medical Image Computing and Computer Assisted Intervention – MICCAI 2023, pp. 263–272, Springer Nature Switzerland, Cham, 2023, ISBN: 978-3-031-43987-2.
@inproceedings{10.1007/978-3-031-43987-2_26,
title = {Detection of Basal Cell Carcinoma in Whole Slide Images},
author = {Hongyan Xu and Dadong Wang and Arcot Sowmya and Ian Katz},
editor = {Hayit Greenspan and Anant Madabhushi and Parvin Mousavi and Septimiu Salcudean and James Duncan and Tanveer Syeda-Mahmood and Russell Taylor},
url = {https://link.springer.com/chapter/10.1007/978-3-031-43987-2_26},
isbn = {978-3-031-43987-2},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Medical Image Computing and Computer Assisted Intervention – MICCAI 2023},
pages = {263–272},
publisher = {Springer Nature Switzerland},
address = {Cham},
abstract = {Basal cell carcinoma (BCC) is a prevalent and increasingly diagnosed form of skin cancer that can benefit from automated whole slide image (WSI) analysis. However, traditional methods that utilize popular network structures designed for natural images, such as the ImageNet dataset, may result in reduced accuracy due to the significant differences between natural and pathology images. In this paper, we analyze skin cancer images using the optimal network obtained by neural architecture search (NAS) on the skin cancer dataset. Compared with traditional methods, our network is more applicable to the task of skin cancer detection. Furthermore, unlike traditional unilaterally augmented (UA) methods, the proposed supernet Skin-Cancer net (SC-net) considers the fairness of training and alleviates the effects of evaluation bias. We use the SC-net to fairly treat all the architectures in the search space and leveraged evolutionary search to obtain the optimal architecture for a skin cancer dataset. Our experiments involve 277,000 patches split from 194 slides. Under the same FLOPs budget (4.1G), our searched ResNet50 model achieves 96.2% accuracy and 96.5% area under the ROC curve (AUC), which are 4.8% and 4.7% higher than those with the baseline settings, respectively.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Basal cell carcinoma (BCC) is a prevalent and increasingly diagnosed form of skin cancer that can benefit from automated whole slide image (WSI) analysis. However, traditional methods that utilize popular network structures designed for natural images, such as the ImageNet dataset, may result in reduced accuracy due to the significant differences between natural and pathology images. In this paper, we analyze skin cancer images using the optimal network obtained by neural architecture search (NAS) on the skin cancer dataset. Compared with traditional methods, our network is more applicable to the task of skin cancer detection. Furthermore, unlike traditional unilaterally augmented (UA) methods, the proposed supernet Skin-Cancer net (SC-net) considers the fairness of training and alleviates the effects of evaluation bias. We use the SC-net to fairly treat all the architectures in the search space and leveraged evolutionary search to obtain the optimal architecture for a skin cancer dataset. Our experiments involve 277,000 patches split from 194 slides. Under the same FLOPs budget (4.1G), our searched ResNet50 model achieves 96.2% accuracy and 96.5% area under the ROC curve (AUC), which are 4.8% and 4.7% higher than those with the baseline settings, respectively.
89.
Yang, Dong; Xu, Ziyue; He, Yufan; Nath, Vishwesh; Li, Wenqi; Myronenko, Andriy; Hatamizadeh, Ali; Zhao, Can; Roth, Holger R.; Xu, Daguang
DAST: Differentiable Architecture Search with Transformer for 3D Medical Image Segmentation Proceedings Article
In: Greenspan, Hayit; Madabhushi, Anant; Mousavi, Parvin; Salcudean, Septimiu; Duncan, James; Syeda-Mahmood, Tanveer; Taylor, Russell (Ed.): Medical Image Computing and Computer Assisted Intervention – MICCAI 2023, pp. 747–756, Springer Nature Switzerland, Cham, 2023, ISBN: 978-3-031-43898-1.
@inproceedings{10.1007/978-3-031-43898-1_71,
title = {DAST: Differentiable Architecture Search with Transformer for 3D Medical Image Segmentation},
author = {Dong Yang and Ziyue Xu and Yufan He and Vishwesh Nath and Wenqi Li and Andriy Myronenko and Ali Hatamizadeh and Can Zhao and Holger R. Roth and Daguang Xu},
editor = {Hayit Greenspan and Anant Madabhushi and Parvin Mousavi and Septimiu Salcudean and James Duncan and Tanveer Syeda-Mahmood and Russell Taylor},
url = {https://link.springer.com/chapter/10.1007/978-3-031-43898-1_71},
isbn = {978-3-031-43898-1},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Medical Image Computing and Computer Assisted Intervention – MICCAI 2023},
pages = {747–756},
publisher = {Springer Nature Switzerland},
address = {Cham},
abstract = {Neural Architecture Search (NAS) has been widely used for medical image segmentation by improving both model performance and computational efficiency. Recently, the Visual Transformer (ViT) model has achieved significant success in computer vision tasks. Leveraging these two innovations, we propose a novel NAS algorithm, DAST, to optimize neural network models with transformers for 3D medical image segmentation. The proposed algorithm is able to search the global structure and local operations of the architecture with a GPU memory consumption constraint. The resulting architectures reveal an effective relationship between convolution and transformer layers in segmentation models. Moreover, we validate the proposed algorithm on large-scale medical image segmentation data sets, showing its superior performance over the baselines. The model achieves state-of-the-art performance in the public challenge of kidney CT segmentation (KiTS'19).},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Neural Architecture Search (NAS) has been widely used for medical image segmentation by improving both model performance and computational efficiency. Recently, the Visual Transformer (ViT) model has achieved significant success in computer vision tasks. Leveraging these two innovations, we propose a novel NAS algorithm, DAST, to optimize neural network models with transformers for 3D medical image segmentation. The proposed algorithm is able to search the global structure and local operations of the architecture with a GPU memory consumption constraint. The resulting architectures reveal an effective relationship between convolution and transformer layers in segmentation models. Moreover, we validate the proposed algorithm on large-scale medical image segmentation data sets, showing its superior performance over the baselines. The model achieves state-of-the-art performance in the public challenge of kidney CT segmentation (KiTS'19).
88.
Ying, Weiqin; Zheng, Qiaoqiao; Wu, Yu; Yang, Kaihao; Zhou, Zekun; Chen, Jiajun; Ye, Zilin
In: Applied Soft Computing, vol. 148, pp. 110869, 2023, ISSN: 1568-4946.
@article{YING2023110869,
title = {Efficient multi-objective evolutionary neural architecture search for U-Nets with diamond atrous convolution and Transformer for medical image segmentation},
author = {Weiqin Ying and Qiaoqiao Zheng and Yu Wu and Kaihao Yang and Zekun Zhou and Jiajun Chen and Zilin Ye},
url = {https://www.sciencedirect.com/science/article/pii/S1568494623008876},
doi = {https://doi.org/10.1016/j.asoc.2023.110869},
issn = {1568-4946},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Applied Soft Computing},
volume = {148},
pages = {110869},
abstract = {Deep encoder–decoder neural networks like U-Nets have made significant contributions to the development of computer vision applications such as image segmentation. Neural architecture search (NAS) has the potential to further automatically adjust the architectures of U-Nets for various medical image segmentation tasks. Most of the NAS techniques focus on optimizing segmentation accuracies of network architectures. In real-world medical image segmentation scenarios, two main challenges are poor medical image quality and diverse deployment devices with different computing capabilities. A large architecture designed only for the high segmentation accuracy is difficult to run on various deployment devices. To address these challenges, this paper proposes a multi-objective evolutionary neural architecture search method (CTU-NAS) for U-Nets with diamond atrous convolution and Transformer for medical image segmentation. A hybrid U-Net architecture (CTU-Net) with diamond atrous convolution and Transformer modules is designed as the supernet of CTU-NAS. Then a channel search strategy based on sorting and selection is applied to speed up the search for subnets by precisely selecting and training the most important channels more times. In addition, CTU-NAS employs a dual acceleration mechanism based on weight sharing and surrogate model to lower the cost of evaluations of subnets. CTU-NAS applies a multi-objective evolutionary algorithm to balance between the segmentation accuracy and the number of parameters. Experimental results on two medical segmentation datasets show that CTU-NAS is capable of quickly generating a group of excellent network architectures with different sizes and their performances also outperform or come close to those of the manually designed networks.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Deep encoder–decoder neural networks like U-Nets have made significant contributions to the development of computer vision applications such as image segmentation. Neural architecture search (NAS) has the potential to further automatically adjust the architectures of U-Nets for various medical image segmentation tasks. Most of the NAS techniques focus on optimizing segmentation accuracies of network architectures. In real-world medical image segmentation scenarios, two main challenges are poor medical image quality and diverse deployment devices with different computing capabilities. A large architecture designed only for the high segmentation accuracy is difficult to run on various deployment devices. To address these challenges, this paper proposes a multi-objective evolutionary neural architecture search method (CTU-NAS) for U-Nets with diamond atrous convolution and Transformer for medical image segmentation. A hybrid U-Net architecture (CTU-Net) with diamond atrous convolution and Transformer modules is designed as the supernet of CTU-NAS. Then a channel search strategy based on sorting and selection is applied to speed up the search for subnets by precisely selecting and training the most important channels more times. In addition, CTU-NAS employs a dual acceleration mechanism based on weight sharing and surrogate model to lower the cost of evaluations of subnets. CTU-NAS applies a multi-objective evolutionary algorithm to balance between the segmentation accuracy and the number of parameters. Experimental results on two medical segmentation datasets show that CTU-NAS is capable of quickly generating a group of excellent network architectures with different sizes and their performances also outperform or come close to those of the manually designed networks.
87.
Kuş, Zeki; Kiraz, Berna
Evolutionary Architecture Optimization for Retinal Vessel Segmentation Journal Article
In: IEEE Journal of Biomedical and Health Informatics, pp. 1-9, 2023.
@article{10250938,
title = {Evolutionary Architecture Optimization for Retinal Vessel Segmentation},
author = {Zeki Kuş and Berna Kiraz},
url = {https://ieeexplore.ieee.org/abstract/document/10250938},
doi = {10.1109/JBHI.2023.3314981},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Journal of Biomedical and Health Informatics},
pages = {1-9},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
86.
Dai, Haixing
Brain-inspired Approaches for Advancing Artificial Intelligence PhD Thesis
University of Georgia, 2023.
@phdthesis{DaiHaixing2023BAfA,
title = {Brain-inspired Approaches for Advancing Artificial Intelligence},
author = {Haixing Dai},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
pages = {197},
school = {University of Georgia},
abstract = {Deep learning has experienced rapid growth and garnered significant attention in recent decades. Simultaneously, neuroscience has remained a challenging and enigmatic field of study. Inspired by the structure and function of the brain, researchers have developed increasingly powerful and sophisticated deep learning models that have achieved remarkable performance in various domains, including computer vision, natural language processing, and medical image analysis. These brain-inspired models have revolutionized the field of artificial intelligence, enabling breakthroughs in tasks such as image recognition, language understanding, and disease diagnosis. In turn, the application of these advanced deep learning models has provided valuable insights into the inner workings of the human brain, revealing temporal and spatial functional brain networks. The symbiotic relationship between artificial intelligence and neuroscience is evident, as they continuously inform and complement each other's progress.This dissertation presents novel frameworks that integrate deep learning and knowledge from brain science. This research aims to gain insights into the brain and refine deep learning models through brain-inspired principles. The dissertation first discusses how deep learning has been applied to study the brain, focusing on areas such as modeling cortical folding patterns, hierarchical brain structures, and spatial-temporal brain networks. It then discusses how artificial neural networks have drawn inspiration from the brain, using examples like convolutional neural networks, attention mechanisms, and language models. The dissertation’s main contributions are several computational frameworks integrating brain-inspired insights. These include a graph representation neural architecture search method to optimize recurrent neural networks for analyzing spatiotemporal brain networks, a hierarchical semantic tree concept whitening model to disentangle concept representations for image classification, a twin-transformer framework to study gyri and sulci in the cortex, a core-periphery guided vision transformer, and methods leveraging language models to generate data and analyze health narratives. Overall, this dissertation explores how we can understand the brain better using deep learning and ultimately build more efficient, robust, and interpretable artificial neural networks inspired by the brain.},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
Deep learning has experienced rapid growth and garnered significant attention in recent decades. Simultaneously, neuroscience has remained a challenging and enigmatic field of study. Inspired by the structure and function of the brain, researchers have developed increasingly powerful and sophisticated deep learning models that have achieved remarkable performance in various domains, including computer vision, natural language processing, and medical image analysis. These brain-inspired models have revolutionized the field of artificial intelligence, enabling breakthroughs in tasks such as image recognition, language understanding, and disease diagnosis. In turn, the application of these advanced deep learning models has provided valuable insights into the inner workings of the human brain, revealing temporal and spatial functional brain networks. The symbiotic relationship between artificial intelligence and neuroscience is evident, as they continuously inform and complement each other's progress.This dissertation presents novel frameworks that integrate deep learning and knowledge from brain science. This research aims to gain insights into the brain and refine deep learning models through brain-inspired principles. The dissertation first discusses how deep learning has been applied to study the brain, focusing on areas such as modeling cortical folding patterns, hierarchical brain structures, and spatial-temporal brain networks. It then discusses how artificial neural networks have drawn inspiration from the brain, using examples like convolutional neural networks, attention mechanisms, and language models. The dissertation’s main contributions are several computational frameworks integrating brain-inspired insights. These include a graph representation neural architecture search method to optimize recurrent neural networks for analyzing spatiotemporal brain networks, a hierarchical semantic tree concept whitening model to disentangle concept representations for image classification, a twin-transformer framework to study gyri and sulci in the cortex, a core-periphery guided vision transformer, and methods leveraging language models to generate data and analyze health narratives. Overall, this dissertation explores how we can understand the brain better using deep learning and ultimately build more efficient, robust, and interpretable artificial neural networks inspired by the brain.
85.
Wu, Jiong; Fan, Yong
HNAS-Reg: Hierarchical Neural Architecture Search for Deformable Medical Image Registration Proceedings Article
In: 20th IEEE International Symposium on Biomedical Imaging, ISBI 2023, Cartagena, Colombia, April 18-21, 2023, pp. 1–4, IEEE, 2023.
@inproceedings{DBLP:conf/isbi/WuF23,
title = {HNAS-Reg: Hierarchical Neural Architecture Search for Deformable Medical Image Registration},
author = {Jiong Wu and Yong Fan},
url = {https://doi.org/10.1109/ISBI53787.2023.10230534},
doi = {10.1109/ISBI53787.2023.10230534},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {20th IEEE International Symposium on Biomedical Imaging, ISBI
2023, Cartagena, Colombia, April 18-21, 2023},
pages = {1–4},
publisher = {IEEE},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
84.
Kuş, Zeki; Kiraz, Berna; Göksu, Tuğçe Koçak; Aydın, Musa; Özkan, Esra; Vural, Atay; Kiraz, Alper; Can, Burhanettin
Differential evolution-based neural architecture search for brain vessel segmentation Journal Article
In: Engineering Science and Technology, an International Journal, vol. 46, pp. 101502, 2023, ISSN: 2215-0986.
@article{KUS2023101502,
title = {Differential evolution-based neural architecture search for brain vessel segmentation},
author = {Zeki Kuş and Berna Kiraz and Tuğçe Koçak Göksu and Musa Aydın and Esra Özkan and Atay Vural and Alper Kiraz and Burhanettin Can},
url = {https://www.sciencedirect.com/science/article/pii/S2215098623001805},
doi = {https://doi.org/10.1016/j.jestch.2023.101502},
issn = {2215-0986},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Engineering Science and Technology, an International Journal},
volume = {46},
pages = {101502},
abstract = {Brain vasculature analysis is critical in developing novel treatment targets for neurodegenerative diseases. Such an accurate analysis cannot be performed manually but requires a semi-automated or fully-automated approach. Deep learning methods have recently proven indispensable for the automated segmentation and analysis of medical images. However, optimizing a deep learning network architecture is another challenge. Manually selecting deep learning network architectures and tuning their hyper-parameters requires a lot of expertise and effort. To solve this problem, neural architecture search (NAS) approaches that explore more efficient network architectures with high segmentation performance have been proposed in the literature. This study introduces differential evolution-based NAS approaches in which a novel search space is proposed for brain vessel segmentation. We select two architectures that are frequently used for medical image segmentation, i.e. U-Net and Attention U-Net, as baselines for NAS optimizations. The conventional differential evolution and the opposition-based differential evolution with novel search space are employed as search methods in NAS. Furthermore, we perform ablation studies and evaluate the effects of specific loss functions, model pruning, threshold selection and generalization performance on the proposed models. The experiments are conducted on two datasets providing 335 single-channel 8-bit gray-scale images. These datasets are a public volumetric cerebrovascular system dataset (vesseINN) and our own dataset called KUVESG. The proposed NAS approaches, namely UNAS-Net and Attention UNAS-Net architectures, yield better segmentation performance in terms of different segmentation metrics. More specifically, UNAS-Net with differential evolution reveals high dice score/sensitivity values of 79.57/81.48, respectively. Moreover, they provide shorter inference times by a factor of 9.15 than the baseline methods.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Brain vasculature analysis is critical in developing novel treatment targets for neurodegenerative diseases. Such an accurate analysis cannot be performed manually but requires a semi-automated or fully-automated approach. Deep learning methods have recently proven indispensable for the automated segmentation and analysis of medical images. However, optimizing a deep learning network architecture is another challenge. Manually selecting deep learning network architectures and tuning their hyper-parameters requires a lot of expertise and effort. To solve this problem, neural architecture search (NAS) approaches that explore more efficient network architectures with high segmentation performance have been proposed in the literature. This study introduces differential evolution-based NAS approaches in which a novel search space is proposed for brain vessel segmentation. We select two architectures that are frequently used for medical image segmentation, i.e. U-Net and Attention U-Net, as baselines for NAS optimizations. The conventional differential evolution and the opposition-based differential evolution with novel search space are employed as search methods in NAS. Furthermore, we perform ablation studies and evaluate the effects of specific loss functions, model pruning, threshold selection and generalization performance on the proposed models. The experiments are conducted on two datasets providing 335 single-channel 8-bit gray-scale images. These datasets are a public volumetric cerebrovascular system dataset (vesseINN) and our own dataset called KUVESG. The proposed NAS approaches, namely UNAS-Net and Attention UNAS-Net architectures, yield better segmentation performance in terms of different segmentation metrics. More specifically, UNAS-Net with differential evolution reveals high dice score/sensitivity values of 79.57/81.48, respectively. Moreover, they provide shorter inference times by a factor of 9.15 than the baseline methods.
83.
Ali, Muhammad Junaid; Moalic, Laurent; Essaid, Mokhtar; Idoumghar, Lhassane
Designing Convolutional Neural Networks Using Surrogate Assisted Genetic Algorithm for Medical Image Classification Proceedings Article
In: Proceedings of the Companion Conference on Genetic and Evolutionary Computation, pp. 263–266, Association for Computing Machinery, Lisbon, Portugal, 2023, ISBN: 9798400701207.
@inproceedings{10.1145/3583133.3590678,
title = {Designing Convolutional Neural Networks Using Surrogate Assisted Genetic Algorithm for Medical Image Classification},
author = {Muhammad Junaid Ali and Laurent Moalic and Mokhtar Essaid and Lhassane Idoumghar},
url = {https://doi.org/10.1145/3583133.3590678},
doi = {10.1145/3583133.3590678},
isbn = {9798400701207},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Proceedings of the Companion Conference on Genetic and Evolutionary Computation},
pages = {263–266},
publisher = {Association for Computing Machinery},
address = {Lisbon, Portugal},
series = {GECCO '23 Companion},
abstract = {Recently, Deep Learning (DL) algorithms have shown state-of-the-art performance in numerous tasks. The design of DL algorithms is time-consuming process that requires expert-level knowledge. To overcome this problem, Neural Architecture Search (NAS) is proposed, which automatically searches for the best neural network architecture for a given task. Indeed, evolutionary NAS has emerged as a widely studied research area, in which evolutionary algorithms are used to design neural networks. In this study, we proposed a novel encoding scheme to represent Convolutional Neural Network (CNN) architectures using continuous representation, which provides a larger search space and potentially finds better solutions compared to discrete representations. Moreover, it allows better exploitation of genetic operators, leading to a quick convergence of individuals. Furthermore, in order to reduce the computational time of the evaluation phase during the search process, surrogate models are used to provide an alternative objective function. To evaluate the effectiveness of the proposal, experiments are performed on CIFAR-10 and multiple datasets from the MedMNIST benchmark. The experimental results demonstrated the effectiveness of the proposed approach compared to several existing state-of-the-art algorithms.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Recently, Deep Learning (DL) algorithms have shown state-of-the-art performance in numerous tasks. The design of DL algorithms is time-consuming process that requires expert-level knowledge. To overcome this problem, Neural Architecture Search (NAS) is proposed, which automatically searches for the best neural network architecture for a given task. Indeed, evolutionary NAS has emerged as a widely studied research area, in which evolutionary algorithms are used to design neural networks. In this study, we proposed a novel encoding scheme to represent Convolutional Neural Network (CNN) architectures using continuous representation, which provides a larger search space and potentially finds better solutions compared to discrete representations. Moreover, it allows better exploitation of genetic operators, leading to a quick convergence of individuals. Furthermore, in order to reduce the computational time of the evaluation phase during the search process, surrogate models are used to provide an alternative objective function. To evaluate the effectiveness of the proposal, experiments are performed on CIFAR-10 and multiple datasets from the MedMNIST benchmark. The experimental results demonstrated the effectiveness of the proposed approach compared to several existing state-of-the-art algorithms.
82.
Carlos, Guilherme; Figueiredo, Karla; Hussain, Abir; Vellasco, Marley
SegQNAS: Quantum-inspired Neural Architecture Search applied to Medical Image Semantic Segmentation Proceedings Article
In: 2023 International Joint Conference on Neural Networks (IJCNN), pp. 1-8, 2023.
@inproceedings{10191869,
title = {SegQNAS: Quantum-inspired Neural Architecture Search applied to Medical Image Semantic Segmentation},
author = {Guilherme Carlos and Karla Figueiredo and Abir Hussain and Marley Vellasco},
doi = {10.1109/IJCNN54540.2023.10191869},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {2023 International Joint Conference on Neural Networks (IJCNN)},
pages = {1-8},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
81.
Qin, Shixi; Zhang, Zixun; Jiang, Yuncheng; Cui, Shuguang; Cheng, Shenghui; Li, Zhen
NG-NAS: Node growth neural architecture search for 3D medical image segmentation Journal Article
In: Computerized Medical Imaging and Graphics, vol. 108, pp. 102268, 2023, ISSN: 0895-6111.
@article{QIN2023102268,
title = {NG-NAS: Node growth neural architecture search for 3D medical image segmentation},
author = {Shixi Qin and Zixun Zhang and Yuncheng Jiang and Shuguang Cui and Shenghui Cheng and Zhen Li},
url = {https://www.sciencedirect.com/science/article/pii/S0895611123000861},
doi = {https://doi.org/10.1016/j.compmedimag.2023.102268},
issn = {0895-6111},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Computerized Medical Imaging and Graphics},
volume = {108},
pages = {102268},
abstract = {Neural architecture search (NAS) has been applied to design proper 3D networks for medical image segmentation. In order to reduce the computation cost in NAS, researchers tend to adopt weight sharing mechanism to search architectures in a supernet. However, recent studies state that the searched architecture rankings may not be accurate with weight sharing mechanism because the training situations are inconsistent between the searching and training phases. In addition, some NAS algorithms design inflexible supernets that only search operators in a pre-defined backbone and ignore the importance of network topology, which limits the performance of searched architecture. To avoid weight sharing mechanism which may lead to inaccurate results and to comprehensively search network topology and operators, we propose a novel NAS algorithm called NG-NAS. Following the previous studies, we consider the segmentation network as a U-shape structure composed of a set of nodes. Instead of searching from the supernet with a limited search space, our NG-NAS starts from a simple architecture with only 5 nodes, and greedily grows the best candidate node until meeting the constraint. We design 2 kinds of node generations to form various network topological structures and prepare 4 candidate operators for each node. To efficiently evaluate candidate node generations, we use NAS without training strategies. We evaluate our method on several public 3D medical image segmentation benchmarks and achieve state-of-the-art performance, demonstrating the effectiveness of the searched architecture and our NG-NAS. Concretely, our method achieves an average Dice score of 85.11 on MSD liver, 65.70 on MSD brain, and 87.59 in BTCV, which performs much better than the previous SOTA methods.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Neural architecture search (NAS) has been applied to design proper 3D networks for medical image segmentation. In order to reduce the computation cost in NAS, researchers tend to adopt weight sharing mechanism to search architectures in a supernet. However, recent studies state that the searched architecture rankings may not be accurate with weight sharing mechanism because the training situations are inconsistent between the searching and training phases. In addition, some NAS algorithms design inflexible supernets that only search operators in a pre-defined backbone and ignore the importance of network topology, which limits the performance of searched architecture. To avoid weight sharing mechanism which may lead to inaccurate results and to comprehensively search network topology and operators, we propose a novel NAS algorithm called NG-NAS. Following the previous studies, we consider the segmentation network as a U-shape structure composed of a set of nodes. Instead of searching from the supernet with a limited search space, our NG-NAS starts from a simple architecture with only 5 nodes, and greedily grows the best candidate node until meeting the constraint. We design 2 kinds of node generations to form various network topological structures and prepare 4 candidate operators for each node. To efficiently evaluate candidate node generations, we use NAS without training strategies. We evaluate our method on several public 3D medical image segmentation benchmarks and achieve state-of-the-art performance, demonstrating the effectiveness of the searched architecture and our NG-NAS. Concretely, our method achieves an average Dice score of 85.11 on MSD liver, 65.70 on MSD brain, and 87.59 in BTCV, which performs much better than the previous SOTA methods.
80.
Jidney, Tasmia Tahmida; Biswas, Angona; Nasim, Md Abdullah Al; Hossain, Ismail; Alam, Md Jahangir; Talukder, Sajedul; Hossain, Mofazzal; Ullah, Md. Azim
AutoML Systems For Medical Imaging Technical Report
2023.
@techreport{DBLP:journals/corr/abs-2306-04750,
title = {AutoML Systems For Medical Imaging},
author = {Tasmia Tahmida Jidney and Angona Biswas and Md Abdullah Al Nasim and Ismail Hossain and Md Jahangir Alam and Sajedul Talukder and Mofazzal Hossain and Md. Azim Ullah},
url = {https://doi.org/10.48550/arXiv.2306.04750},
doi = {10.48550/arXiv.2306.04750},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {CoRR},
volume = {abs/2306.04750},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
79.
Ye, Shaozhuang; Wang, Tuo; Ding, Mingyue; Zhang, Xuming
F-DARTS: Foveated Differentiable Architecture Search Based Multimodal Medical Image Fusion Journal Article
In: IEEE Transactions on Medical Imaging, pp. 1-1, 2023.
@article{10145413,
title = {F-DARTS: Foveated Differentiable Architecture Search Based Multimodal Medical Image Fusion},
author = {Shaozhuang Ye and Tuo Wang and Mingyue Ding and Xuming Zhang},
url = {https://ieeexplore.ieee.org/abstract/document/10145413},
doi = {10.1109/TMI.2023.3283517},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Transactions on Medical Imaging},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
78.
Alaiad, Ahmad; Migdady, Aya; Al-Khatib, Ra’ed M.; Alzoubi, Omar; Zitar, Raed Abu; Abualigah, Laith
Autokeras Approach: A Robust Automated Deep Learning Network for Diagnosis Disease Cases in Medical Images Journal Article
In: Journal of Imaging, vol. 9, no. 3, 2023, ISSN: 2313-433X.
@article{jimaging9030064,
title = {Autokeras Approach: A Robust Automated Deep Learning Network for Diagnosis Disease Cases in Medical Images},
author = {Ahmad Alaiad and Aya Migdady and Ra’ed M. Al-Khatib and Omar Alzoubi and Raed Abu Zitar and Laith Abualigah},
url = {https://www.mdpi.com/2313-433X/9/3/64},
doi = {10.3390/jimaging9030064},
issn = {2313-433X},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Journal of Imaging},
volume = {9},
number = {3},
abstract = {Automated deep learning is promising in artificial intelligence (AI). However, a few applications of automated deep learning networks have been made in the clinical medical fields. Therefore, we studied the application of an open-source automated deep learning framework, Autokeras, for detecting smear blood images infected with malaria parasites. Autokeras is able to identify the optimal neural network to perform the classification task. Hence, the robustness of the adopted model is due to it not needing any prior knowledge from deep learning. In contrast, the traditional deep neural network methods still require more construction to identify the best convolutional neural network (CNN). The dataset used in this study consisted of 27,558 blood smear images. A comparative process proved the superiority of our proposed approach over other traditional neural networks. The evaluation results of our proposed model achieved high efficiency with impressive accuracy, reaching 95.6% when compared with previous competitive models.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Automated deep learning is promising in artificial intelligence (AI). However, a few applications of automated deep learning networks have been made in the clinical medical fields. Therefore, we studied the application of an open-source automated deep learning framework, Autokeras, for detecting smear blood images infected with malaria parasites. Autokeras is able to identify the optimal neural network to perform the classification task. Hence, the robustness of the adopted model is due to it not needing any prior knowledge from deep learning. In contrast, the traditional deep neural network methods still require more construction to identify the best convolutional neural network (CNN). The dataset used in this study consisted of 27,558 blood smear images. A comparative process proved the superiority of our proposed approach over other traditional neural networks. The evaluation results of our proposed model achieved high efficiency with impressive accuracy, reaching 95.6% when compared with previous competitive models.
77.
Rajesh, Chilukamari; Kumar, Sushil
Äutomatic Retinal Vessel Segmentation Using BTLBO Proceedings Article
In: Thakur, Manoj; Agnihotri, Samar; Rajpurohit, Bharat Singh; Pant, Millie; Deep, Kusum; Nagar, Atulya K. (Ed.): Soft Computing for Problem Solving, pp. 189–200, Springer Nature Singapore, Singapore, 2023, ISBN: 978-981-19-6525-8.
@inproceedings{10.1007/978-981-19-6525-8_15,
title = {Äutomatic Retinal Vessel Segmentation Using BTLBO},
author = {Chilukamari Rajesh and Sushil Kumar},
editor = {Manoj Thakur and Samar Agnihotri and Bharat Singh Rajpurohit and Millie Pant and Kusum Deep and Atulya K. Nagar},
url = {https://link.springer.com/chapter/10.1007/978-981-19-6525-8_15},
isbn = {978-981-19-6525-8},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Soft Computing for Problem Solving},
pages = {189--200},
publisher = {Springer Nature Singapore},
address = {Singapore},
abstract = {The accuracy of retinal vessel segmentation (RVS) is crucial in assisting physicians in the ophthalmology diagnosis or other systemic diseases. However, manual segmentation needs a high level of knowledge, time-consuming, complex, and prone to errors. As a result, automatic vessel segmentation is required, which might be a significant technological breakthrough in the medical field. We proposed a novel strategy in this paper, that uses neural architecture search (NAS) to optimize a U-net architecture using a binary teaching learning-based optimization (BTLBO) evolutionary algorithm for RVS to increase vessel segmentation performance and reduce the workload of manually developing deep networks with limited computing resources. We used a publicly available DRIVE dataset to examine the proposed approach and showed that the discovered model generated by the proposed approach outperforms existing methods.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The accuracy of retinal vessel segmentation (RVS) is crucial in assisting physicians in the ophthalmology diagnosis or other systemic diseases. However, manual segmentation needs a high level of knowledge, time-consuming, complex, and prone to errors. As a result, automatic vessel segmentation is required, which might be a significant technological breakthrough in the medical field. We proposed a novel strategy in this paper, that uses neural architecture search (NAS) to optimize a U-net architecture using a binary teaching learning-based optimization (BTLBO) evolutionary algorithm for RVS to increase vessel segmentation performance and reduce the workload of manually developing deep networks with limited computing resources. We used a publicly available DRIVE dataset to examine the proposed approach and showed that the discovered model generated by the proposed approach outperforms existing methods.
76.
Kuş, Zeki; Aydin, Musa; Kiraz, Berna; Can, Burhanettin
Neural Architecture Search Using Metaheuristics for Automated Cell Segmentation Proceedings Article
In: Gaspero, Luca Di; Festa, Paola; Nakib, Amir; Pavone, Mario (Ed.): Metaheuristics, pp. 158–171, Springer International Publishing, Cham, 2023, ISBN: 978-3-031-26504-4.
@inproceedings{10.1007/978-3-031-26504-4_12,
title = {Neural Architecture Search Using Metaheuristics for Automated Cell Segmentation},
author = {Zeki Kuş and Musa Aydin and Berna Kiraz and Burhanettin Can},
editor = {Luca Di Gaspero and Paola Festa and Amir Nakib and Mario Pavone},
url = {https://link.springer.com/chapter/10.1007/978-3-031-26504-4_12},
isbn = {978-3-031-26504-4},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {Metaheuristics},
pages = {158--171},
publisher = {Springer International Publishing},
address = {Cham},
abstract = {Deep neural networks give successful results for segmentation of medical images. The need for optimizing many hyper-parameters presents itself as a significant limitation hampering the effectiveness of deep neural network based segmentation task. Manual selection of these hyper-parameters is not feasible as the search space increases. At the same time, these generated networks are problem-specific. Recently, studies that perform segmentation of medical images using Neural Architecture Search (NAS) have been proposed. However, these studies significantly limit the possible network structures and search space. In this study, we proposed a structure called UNAS-Net that brings together the advantages of successful NAS studies and is more flexible in terms of the networks that can be created. The UNAS-Net structure has been optimized using metaheuristics including Differential Evolution (DE) and Local Search (LS), and the generated networks have been tested on Optofil and Cell Nuclei data sets. When the results are examined, it is seen that the networks produced by the heuristic methods improve the performance of the U-Net structure in terms of both segmentation performance and computational complexity. As a result, the proposed structure can be used when the automatic generation of neural networks that provide fast inference as well as successful segmentation performance is desired.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Deep neural networks give successful results for segmentation of medical images. The need for optimizing many hyper-parameters presents itself as a significant limitation hampering the effectiveness of deep neural network based segmentation task. Manual selection of these hyper-parameters is not feasible as the search space increases. At the same time, these generated networks are problem-specific. Recently, studies that perform segmentation of medical images using Neural Architecture Search (NAS) have been proposed. However, these studies significantly limit the possible network structures and search space. In this study, we proposed a structure called UNAS-Net that brings together the advantages of successful NAS studies and is more flexible in terms of the networks that can be created. The UNAS-Net structure has been optimized using metaheuristics including Differential Evolution (DE) and Local Search (LS), and the generated networks have been tested on Optofil and Cell Nuclei data sets. When the results are examined, it is seen that the networks produced by the heuristic methods improve the performance of the U-Net structure in terms of both segmentation performance and computational complexity. As a result, the proposed structure can be used when the automatic generation of neural networks that provide fast inference as well as successful segmentation performance is desired.
75.
Eminaga, Okyaz; Abbas, Mahmoud; Shen, Jeanne; Laurie, Mark; Brooks, James D.; Liao, Joseph C.; Rubin, Daniel L.
PlexusNet: A neural network architectural concept for medical image classification Journal Article
In: Computers in Biology and Medicine, pp. 106594, 2023, ISSN: 0010-4825.
@article{EMINAGA2023106594,
title = {PlexusNet: A neural network architectural concept for medical image classification},
author = {Okyaz Eminaga and Mahmoud Abbas and Jeanne Shen and Mark Laurie and James D. Brooks and Joseph C. Liao and Daniel L. Rubin},
url = {https://www.sciencedirect.com/science/article/pii/S0010482523000598},
doi = {https://doi.org/10.1016/j.compbiomed.2023.106594},
issn = {0010-4825},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Computers in Biology and Medicine},
pages = {106594},
abstract = {State-of-the-art (SOTA) convolutional neural network models have been widely adapted in medical imaging and applied to address different clinical problems. However, the complexity and scale of such models may not be justified in medical imaging and subject to the available resource budget. Further increasing the number of representative feature maps for the classification task decreases the model explainability. The current data normalization practice is fixed prior to model development and discounting the specification of the data domain. Acknowledging these issues, the current work proposed a new scalable model family called PlexusNet; the block architecture and model scaling by the network's depth, width, and branch regulate PlexusNet's architecture. The efficient computation costs outlined the dimensions of PlexusNet scaling and design. PlexusNet includes a new learnable data normalization algorithm for better data generalization. We applied a simple yet effective neural architecture search to design PlexusNet tailored to five clinical classification problems that achieve a performance noninferior to the SOTA models ResNet-18 and EfficientNet B0/1. It also does so with lower parameter capacity and representative feature maps in ten-fold ranges than the smallest SOTA models with comparable performance. The visualization of representative features revealed distinguishable clusters associated with categories based on latent features generated by PlexusNet. The package and source code are at https://github.com/oeminaga/PlexusNet.git.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
State-of-the-art (SOTA) convolutional neural network models have been widely adapted in medical imaging and applied to address different clinical problems. However, the complexity and scale of such models may not be justified in medical imaging and subject to the available resource budget. Further increasing the number of representative feature maps for the classification task decreases the model explainability. The current data normalization practice is fixed prior to model development and discounting the specification of the data domain. Acknowledging these issues, the current work proposed a new scalable model family called PlexusNet; the block architecture and model scaling by the network's depth, width, and branch regulate PlexusNet's architecture. The efficient computation costs outlined the dimensions of PlexusNet scaling and design. PlexusNet includes a new learnable data normalization algorithm for better data generalization. We applied a simple yet effective neural architecture search to design PlexusNet tailored to five clinical classification problems that achieve a performance noninferior to the SOTA models ResNet-18 and EfficientNet B0/1. It also does so with lower parameter capacity and representative feature maps in ten-fold ranges than the smallest SOTA models with comparable performance. The visualization of representative features revealed distinguishable clusters associated with categories based on latent features generated by PlexusNet. The package and source code are at https://github.com/oeminaga/PlexusNet.git.
74.
Saeed, Fahman; Hussain, Muhammad; Aboalsamh, Hatim A.; Adel, Fadwa Al; Owaifeer, Adi Mohammed Al
Designing the Architecture of a Convolutional Neural Network Automatically for Diabetic Retinopathy Diagnosis Journal Article
In: Mathematics, vol. 11, no. 2, 2023, ISSN: 2227-7390.
@article{math11020307,
title = {Designing the Architecture of a Convolutional Neural Network Automatically for Diabetic Retinopathy Diagnosis},
author = {Fahman Saeed and Muhammad Hussain and Hatim A. Aboalsamh and Fadwa Al Adel and Adi Mohammed Al Owaifeer},
url = {https://www.mdpi.com/2227-7390/11/2/307},
doi = {10.3390/math11020307},
issn = {2227-7390},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {Mathematics},
volume = {11},
number = {2},
abstract = {Diabetic retinopathy (DR) is a leading cause of blindness in middle-aged diabetic patients. Regular screening for DR using fundus imaging aids in detecting complications and delays the progression of the disease. Because manual screening takes time and is subjective, deep learning has been used to help graders. Pre-trained or brute force CNN models are used in existing DR grading CNN-based approaches that are not suited to fundus image complexity. To solve this problem, we present a method for automatically customizing CNN models based on fundus image lesions. It uses k-medoid clustering, principal component analysis (PCA), and inter-class and intra-class variations to determine the CNN model’s depth and width. The designed models are lightweight, adapted to the internal structures of fundus images, and encode the discriminative patterns of DR lesions. The technique is validated on a local dataset from King Saud University Medical City, Saudi Arabia, and two challenging Kaggle datasets: EyePACS and APTOS2019. The auto-designed models outperform well-known pre-trained CNN models such as ResNet152, DenseNet121, and ResNeSt50, as well as Google’s AutoML and Auto-Keras models based on neural architecture search (NAS). The proposed method outperforms current CNN-based DR screening methods. The proposed method can be used in various clinical settings to screen for DR and refer patients to ophthalmologists for further evaluation and treatment.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Diabetic retinopathy (DR) is a leading cause of blindness in middle-aged diabetic patients. Regular screening for DR using fundus imaging aids in detecting complications and delays the progression of the disease. Because manual screening takes time and is subjective, deep learning has been used to help graders. Pre-trained or brute force CNN models are used in existing DR grading CNN-based approaches that are not suited to fundus image complexity. To solve this problem, we present a method for automatically customizing CNN models based on fundus image lesions. It uses k-medoid clustering, principal component analysis (PCA), and inter-class and intra-class variations to determine the CNN model’s depth and width. The designed models are lightweight, adapted to the internal structures of fundus images, and encode the discriminative patterns of DR lesions. The technique is validated on a local dataset from King Saud University Medical City, Saudi Arabia, and two challenging Kaggle datasets: EyePACS and APTOS2019. The auto-designed models outperform well-known pre-trained CNN models such as ResNet152, DenseNet121, and ResNeSt50, as well as Google’s AutoML and Auto-Keras models based on neural architecture search (NAS). The proposed method outperforms current CNN-based DR screening methods. The proposed method can be used in various clinical settings to screen for DR and refer patients to ophthalmologists for further evaluation and treatment.
2022
73.
Cui, S.; Wang, J.; Gui, X.; Wang, T.; Ma, F.
AUTOMED: Automated Medical Risk Predictive Modeling on Electronic Health Records Proceedings Article
In: 2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), pp. 948-953, IEEE Computer Society, Los Alamitos, CA, USA, 2022.
@inproceedings{9995209,
title = {AUTOMED: Automated Medical Risk Predictive Modeling on Electronic Health Records},
author = {S. Cui and J. Wang and X. Gui and T. Wang and F. Ma},
url = {https://doi.ieeecomputersociety.org/10.1109/BIBM55620.2022.9995209},
doi = {10.1109/BIBM55620.2022.9995209},
year = {2022},
date = {2022-12-01},
urldate = {2022-12-01},
booktitle = {2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)},
pages = {948-953},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
abstract = {Electronic health records (EHR) have been widely applied to various tasks in the medical domain such as risk predictive modeling, which aims to predict further health conditions by analyzing patients’ historical EHR. Existing work mainly focuses on modeling the sequential and temporal characteristics of EHR data with advanced deep learning techniques. However, the network architectures of these models are all manually designed based on experts’ prior knowledge, which largely impedes non-experts from exploring this task. To address this issue, in this paper, we propose a novel automated risk prediction model named AUTOMED to automatically search the optimal model architecture for modeling the complex EHR data and improving the performance of the risk prediction task. In particular, we follow the idea of neural architecture search to design a search space that contains three separate searchable modules. Two of them are used for analyzing sequential and temporal features of EHR data, respectively. The third is to automatically fuse both features together. Besides these three modules, AUTOMED contains an embedding module and a prediction module. All the three searchable modules are jointly optimized in the search stage to derive the optimal model architecture. In such a way, the model design can be automatically achieved with few human interventions. Experimental results on three real-world datasets show that AUTOMED outperforms state-of-the-art baselines in terms of PR-AUC, F1, and Cohen’s Kappa. Moreover, the ablation study shows that AUTOMED can obtain reasonable model architectures and offer useful insights to the future risk prediction model design.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Electronic health records (EHR) have been widely applied to various tasks in the medical domain such as risk predictive modeling, which aims to predict further health conditions by analyzing patients’ historical EHR. Existing work mainly focuses on modeling the sequential and temporal characteristics of EHR data with advanced deep learning techniques. However, the network architectures of these models are all manually designed based on experts’ prior knowledge, which largely impedes non-experts from exploring this task. To address this issue, in this paper, we propose a novel automated risk prediction model named AUTOMED to automatically search the optimal model architecture for modeling the complex EHR data and improving the performance of the risk prediction task. In particular, we follow the idea of neural architecture search to design a search space that contains three separate searchable modules. Two of them are used for analyzing sequential and temporal features of EHR data, respectively. The third is to automatically fuse both features together. Besides these three modules, AUTOMED contains an embedding module and a prediction module. All the three searchable modules are jointly optimized in the search stage to derive the optimal model architecture. In such a way, the model design can be automatically achieved with few human interventions. Experimental results on three real-world datasets show that AUTOMED outperforms state-of-the-art baselines in terms of PR-AUC, F1, and Cohen’s Kappa. Moreover, the ablation study shows that AUTOMED can obtain reasonable model architectures and offer useful insights to the future risk prediction model design.
72.
Lu, Zexin; Xia, Wenjun; Huang, Yongqiang; Hou, Mingzheng; Chen, Hu; Zhou, Jiliu; Shan, Hongming; Zhang, Yi
M<sup>3</sup>NAS: Multi-Scale and Multi-Level Memory-Efficient Neural Architecture Search for Low-Dose CT Denoising Journal Article
In: IEEE transactions on medical imaging, vol. PP, 2022, ISSN: 0278-0062.
@article{PMID:36327187,
title = {M^{3}NAS: Multi-Scale and Multi-Level Memory-Efficient Neural Architecture Search for Low-Dose CT Denoising},
author = {Zexin Lu and Wenjun Xia and Yongqiang Huang and Mingzheng Hou and Hu Chen and Jiliu Zhou and Hongming Shan and Yi Zhang},
url = {https://doi.org/10.1109/TMI.2022.3219286},
doi = {10.1109/tmi.2022.3219286},
issn = {0278-0062},
year = {2022},
date = {2022-11-01},
urldate = {2022-11-01},
journal = {IEEE transactions on medical imaging},
volume = {PP},
abstract = {Lowering the radiation dose in computed tomography (CT) can greatly reduce the potential risk to public health. However, the reconstructed images from dose-reduced CT or low-dose CT (LDCT) suffer from severe noise which compromises the subsequent diagnosis and analysis. Recently, convolutional neural networks have achieved promising results in removing noise from LDCT images. The network architectures that are used are either handcrafted or built on top of conventional networks such as ResNet and U-Net. Recent advances in neural network architecture search (NAS) have shown that the network architecture has a dramatic effect on the model performance. This indicates that current network architectures for LDCT may be suboptimal. Therefore, in this paper, we make the first attempt to apply NAS to LDCT and propose a multi-scale and multi-level memory-efficient NAS for LDCT denoising, termed M^{3}NAS. On the one hand, the proposed M^{3}NAS fuses features extracted by different scale cells to capture multi-scale image structural details. On the other hand, the proposed M^{3}NAS can search a hybrid cell- and network-level structure for better performance. In addition, M^{3}NAS can effectively reduce the number of model parameters and increase the speed of inference. Extensive experimental results on two different datasets demonstrate that the proposed M^{3}NAS can achieve better performance and fewer parameters than several state-of-the-art methods. In addition, we also validate the effectiveness of the multi-scale and multi-level architecture for LDCT denoising, and present further analysis for different configurations of super-net.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Lowering the radiation dose in computed tomography (CT) can greatly reduce the potential risk to public health. However, the reconstructed images from dose-reduced CT or low-dose CT (LDCT) suffer from severe noise which compromises the subsequent diagnosis and analysis. Recently, convolutional neural networks have achieved promising results in removing noise from LDCT images. The network architectures that are used are either handcrafted or built on top of conventional networks such as ResNet and U-Net. Recent advances in neural network architecture search (NAS) have shown that the network architecture has a dramatic effect on the model performance. This indicates that current network architectures for LDCT may be suboptimal. Therefore, in this paper, we make the first attempt to apply NAS to LDCT and propose a multi-scale and multi-level memory-efficient NAS for LDCT denoising, termed M<sup>3</sup>NAS. On the one hand, the proposed M<sup>3</sup>NAS fuses features extracted by different scale cells to capture multi-scale image structural details. On the other hand, the proposed M<sup>3</sup>NAS can search a hybrid cell- and network-level structure for better performance. In addition, M<sup>3</sup>NAS can effectively reduce the number of model parameters and increase the speed of inference. Extensive experimental results on two different datasets demonstrate that the proposed M<sup>3</sup>NAS can achieve better performance and fewer parameters than several state-of-the-art methods. In addition, we also validate the effectiveness of the multi-scale and multi-level architecture for LDCT denoising, and present further analysis for different configurations of super-net.
71.
Fakieh, Bahjat; Ragab, Mahmoud
Automated COVID-19 Classification Using Heap-Based Optimization with the Deep Transfer Learning Model Journal Article Forthcoming
In: Computational Intelligence and its Applications in Biomedical Engineering, vol. 2022, Forthcoming.
@article{FakiehCIN2022,
title = {Automated COVID-19 Classification Using Heap-Based Optimization with the Deep Transfer Learning Model},
author = {Bahjat Fakieh and Mahmoud Ragab},
url = {https://www.hindawi.com/journals/cin/2022/7508836/},
doi = { https://doi.org/10.1155/2022/7508836},
year = {2022},
date = {2022-08-01},
urldate = {2022-08-01},
journal = {Computational Intelligence and its Applications in Biomedical Engineering},
volume = {2022},
keywords = {},
pubstate = {forthcoming},
tppubtype = {article}
}
70.
Li, Jolen; Galazis, Christopher; Popov, Illarion; Ovchinnikov, Lev; Vesnin, Sergey; Losev, Alexander; Goryanin, Igor
Dynamic Weight Agnostic Neural Networks and Medical Microwave Radiometry (MWR) for Breast Cancer Diagnostics Technical Report
2022.
@techreport{LiPrePrint2022,
title = {Dynamic Weight Agnostic Neural Networks and Medical Microwave Radiometry (MWR) for Breast Cancer Diagnostics},
author = { Jolen Li and Christopher Galazis and Illarion Popov and Lev Ovchinnikov and Sergey Vesnin and Alexander Losev and Igor Goryanin},
url = {https://www.preprints.org/manuscript/202207.0370/v1},
doi = {10.20944/preprints202207.0370.v1},
year = {2022},
date = {2022-07-29},
urldate = {2022-07-29},
howpublished = {Preprints 2022},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
69.
AL-Ghamdi, Abdullah S. AL-Malaise; Ragab, Mahmoud; AlGhamdi, Saad Abdulla; Asseri, Amer H.; ans Deepika Koundal, Romany F. Mansour
Cognitive Computing Paradigms for Medical Big Data Processing and its Trends Journal Article
In: Computational Intelligence and Neuroscience, 2022.
@article{GhamdiCCP2022,
title = {Cognitive Computing Paradigms for Medical Big Data Processing and its Trends},
author = {Abdullah S. AL-Malaise AL-Ghamdi and Mahmoud Ragab and Saad Abdulla AlGhamdi and Amer H. Asseri and Romany F. Mansour ans Deepika Koundal},
url = {https://www.hindawi.com/journals/cin/2022/3500552/},
year = {2022},
date = {2022-04-30},
urldate = {2022-04-30},
journal = {Computational Intelligence and Neuroscience},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
68.
Dai, Haixing; Li, Qing; Zhao, Lin; Pan, Liming; Shi, Cheng; Liu, Zhengliang; Wu, Zihao; Zhang, Lu; Zhao, Shijie; Wu, Xia; Liu, Tianming; Zhu, Dajiang
Graph Representation Neural Architecture Search for Optimal Spatial/Temporal Functional Brain Network Decomposition Proceedings Article
In: Lian, Chunfeng; Cao, Xiaohuan; Rekik, Islem; Xu, Xuanang; Cui, Zhiming (Ed.): Machine Learning in Medical Imaging, pp. 279–287, Springer Nature Switzerland, Cham, 2022, ISBN: 978-3-031-21014-3.
@inproceedings{10.1007/978-3-031-21014-3_29,
title = {Graph Representation Neural Architecture Search for Optimal Spatial/Temporal Functional Brain Network Decomposition},
author = {Haixing Dai and Qing Li and Lin Zhao and Liming Pan and Cheng Shi and Zhengliang Liu and Zihao Wu and Lu Zhang and Shijie Zhao and Xia Wu and Tianming Liu and Dajiang Zhu},
editor = {Chunfeng Lian and Xiaohuan Cao and Islem Rekik and Xuanang Xu and Zhiming Cui},
url = {https://link.springer.com/chapter/10.1007/978-3-031-21014-3_29},
isbn = {978-3-031-21014-3},
year = {2022},
date = {2022-01-01},
urldate = {2022-01-01},
booktitle = {Machine Learning in Medical Imaging},
pages = {279--287},
publisher = {Springer Nature Switzerland},
address = {Cham},
abstract = {Decomposing the spatial/temporal functional brain networks from 4D functional magnetic resonance imaging (fMRI) data has attracted extensive attention. Among all these efforts, deep neural network-based methods have shown significant advantages due to their powerful hierarchical representation ability. However, the network architectures of those deep learning models are manually crafted, which is time consuming and non-optimal. This paper presents a novel graph representation neural architecture search (GR-NAS) method based on graph representation to optimize the vanilla RNN cell structure for decomposing spatial/temporal brain networks. The core idea is to embed the discrete search space of the RNN cell into a continuous domain that preserves the topological information. After that, popular search algorithms, e.g., reinforcement learning (RL) and Bayesian optimization (BO), can be employed to find the optimal architecture in this continuous space. The proposed method was evaluated on the Human Connectome Project (HCP) task fMRI datasets. Extensive experiments demonstrated the superiority of the proposed model in brain network decomposition both spatially and temporally. To our best knowledge, the proposed model is among the early efforts using NAS strategy to optimally decompose spatial/temporal functional brain networks from fMRI data.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Decomposing the spatial/temporal functional brain networks from 4D functional magnetic resonance imaging (fMRI) data has attracted extensive attention. Among all these efforts, deep neural network-based methods have shown significant advantages due to their powerful hierarchical representation ability. However, the network architectures of those deep learning models are manually crafted, which is time consuming and non-optimal. This paper presents a novel graph representation neural architecture search (GR-NAS) method based on graph representation to optimize the vanilla RNN cell structure for decomposing spatial/temporal brain networks. The core idea is to embed the discrete search space of the RNN cell into a continuous domain that preserves the topological information. After that, popular search algorithms, e.g., reinforcement learning (RL) and Bayesian optimization (BO), can be employed to find the optimal architecture in this continuous space. The proposed method was evaluated on the Human Connectome Project (HCP) task fMRI datasets. Extensive experiments demonstrated the superiority of the proposed model in brain network decomposition both spatially and temporally. To our best knowledge, the proposed model is among the early efforts using NAS strategy to optimally decompose spatial/temporal functional brain networks from fMRI data.
67.
Shen, Jinbo; Luo, Mengting; Liu, Han; Liao, Peixi; Chen, Hu; Zhang, Yi
MLF-IOSC: Multi-Level Fusion Network with Independent Operation Search Cell for Low-Dose CT Denoising Journal Article
In: IEEE Transactions on Medical Imaging, pp. 1-1, 2022.
@article{9963565,
title = {MLF-IOSC: Multi-Level Fusion Network with Independent Operation Search Cell for Low-Dose CT Denoising},
author = {Jinbo Shen and Mengting Luo and Han Liu and Peixi Liao and Hu Chen and Yi Zhang},
url = {https://ieeexplore.ieee.org/abstract/document/9963565},
doi = {10.1109/TMI.2022.3224396},
year = {2022},
date = {2022-01-01},
urldate = {2022-01-01},
journal = {IEEE Transactions on Medical Imaging},
pages = {1-1},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
66.
Hu, Xiaobin; Shen, Ruolin; Luo, Donghao; Tai, Ying; Wang, Chengjie; Menze, Bjoern H.
AutoGAN-Synthesizer: Neural Architecture Search for Cross-Modality MRI Synthesis Proceedings Article
In: Wang, Linwei; Dou, Qi; Fletcher, P. Thomas; Speidel, Stefanie; Li, Shuo (Ed.): Medical Image Computing and Computer Assisted Intervention -- MICCAI 2022, pp. 397–409, Springer Nature Switzerland, Cham, 2022, ISBN: 978-3-031-16446-0.
@inproceedings{10.1007/978-3-031-16446-0_38,
title = {AutoGAN-Synthesizer: Neural Architecture Search for Cross-Modality MRI Synthesis},
author = {Xiaobin Hu and Ruolin Shen and Donghao Luo and Ying Tai and Chengjie Wang and Bjoern H. Menze},
editor = {Linwei Wang and Qi Dou and P. Thomas Fletcher and Stefanie Speidel and Shuo Li},
url = {https://link.springer.com/chapter/10.1007/978-3-031-16446-0_38},
isbn = {978-3-031-16446-0},
year = {2022},
date = {2022-01-01},
urldate = {2022-01-01},
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2022},
pages = {397--409},
publisher = {Springer Nature Switzerland},
address = {Cham},
abstract = {Considering the difficulty to obtain complete multi-modality MRI scans in some real-world data acquisition situations, synthesizing MRI data is a highly relevant and important topic to complement diagnosis information in clinical practice. In this study, we present a novel MRI synthesizer, called AutoGAN-Synthesizer, which automatically discovers generative networks for cross-modality MRI synthesis. Our AutoGAN-Synthesizer adopts gradient-based search strategies to explore the generator architecture by determining how to fuse multi-resolution features and utilizes GAN-based perceptual searching losses to handle the trade-off between model complexity and performance. Our AutoGAN-Synthesizer can search for a remarkable and light-weight architecture with 6.31 Mb parameters only occupying 12 GPU hours. Moreover, to incorporate richer prior knowledge for MRI synthesis, we derive K-space features containing the low- and high-spatial frequency information and incorporate such features into our model. To our best knowledge, this is the first work to explore AutoML for cross-modality MRI synthesis, and our approach is also capable of tailoring networks given either different multiple modalities or just a single modality as input. Extensive experiments show that our AutoGAN-Synthesizer outperforms the state-of-the-art MRI synthesis methods both quantitatively and qualitatively. The code are available at https://github.com/HUuxiaobin/AutoGAN-Synthesizer.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Considering the difficulty to obtain complete multi-modality MRI scans in some real-world data acquisition situations, synthesizing MRI data is a highly relevant and important topic to complement diagnosis information in clinical practice. In this study, we present a novel MRI synthesizer, called AutoGAN-Synthesizer, which automatically discovers generative networks for cross-modality MRI synthesis. Our AutoGAN-Synthesizer adopts gradient-based search strategies to explore the generator architecture by determining how to fuse multi-resolution features and utilizes GAN-based perceptual searching losses to handle the trade-off between model complexity and performance. Our AutoGAN-Synthesizer can search for a remarkable and light-weight architecture with 6.31 Mb parameters only occupying 12 GPU hours. Moreover, to incorporate richer prior knowledge for MRI synthesis, we derive K-space features containing the low- and high-spatial frequency information and incorporate such features into our model. To our best knowledge, this is the first work to explore AutoML for cross-modality MRI synthesis, and our approach is also capable of tailoring networks given either different multiple modalities or just a single modality as input. Extensive experiments show that our AutoGAN-Synthesizer outperforms the state-of-the-art MRI synthesis methods both quantitatively and qualitatively. The code are available at https://github.com/HUuxiaobin/AutoGAN-Synthesizer.
65.
He, Xin; Ying, Guohao; Zhang, Jiyong; Chu, Xiaowen
Evolutionary Multi-objective Architecture Search Framework: Application to COVID-19 3D CT Classification Proceedings Article
In: Wang, Linwei; Dou, Qi; Fletcher, P. Thomas; Speidel, Stefanie; Li, Shuo (Ed.): Medical Image Computing and Computer Assisted Intervention -- MICCAI 2022, pp. 560–570, Springer Nature Switzerland, Cham, 2022, ISBN: 978-3-031-16431-6.
@inproceedings{10.1007/978-3-031-16431-6_53,
title = {Evolutionary Multi-objective Architecture Search Framework: Application to COVID-19 3D CT Classification},
author = {Xin He and Guohao Ying and Jiyong Zhang and Xiaowen Chu},
editor = {Linwei Wang and Qi Dou and P. Thomas Fletcher and Stefanie Speidel and Shuo Li},
isbn = {978-3-031-16431-6},
year = {2022},
date = {2022-01-01},
urldate = {2022-01-01},
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2022},
pages = {560--570},
publisher = {Springer Nature Switzerland},
address = {Cham},
abstract = {The COVID-19 pandemic has threatened global health. Many studies have applied deep convolutional neural networks (CNN) to recognize COVID-19 based on chest 3D computed tomography (CT). Recent works show that no model generalizes well across CT datasets from different countries, and manually designing models for specific datasets requires expertise; thus, neural architecture search (NAS) that aims to search models automatically has become an attractive solution. To reduce the search cost on large 3D CT datasets, most NAS-based works use the weight-sharing (WS) strategy to make all models share weights within a supernet; however, WS inevitably incurs search instability, leading to inaccurate model estimation. In this work, we propose an efficient Evolutionary Multi-objective ARchitecture Search (EMARS) framework. We propose a new objective, namely potential, which can help exploit promising models to indirectly reduce the number of models involved in weights training, thus alleviating search instability. We demonstrate that under objectives of accuracy and potential, EMARS can balance exploitation and exploration, i.e., reducing search time and finding better models. Our searched models are small and perform better than prior works on three public COVID-19 3D CT datasets.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The COVID-19 pandemic has threatened global health. Many studies have applied deep convolutional neural networks (CNN) to recognize COVID-19 based on chest 3D computed tomography (CT). Recent works show that no model generalizes well across CT datasets from different countries, and manually designing models for specific datasets requires expertise; thus, neural architecture search (NAS) that aims to search models automatically has become an attractive solution. To reduce the search cost on large 3D CT datasets, most NAS-based works use the weight-sharing (WS) strategy to make all models share weights within a supernet; however, WS inevitably incurs search instability, leading to inaccurate model estimation. In this work, we propose an efficient Evolutionary Multi-objective ARchitecture Search (EMARS) framework. We propose a new objective, namely potential, which can help exploit promising models to indirectly reduce the number of models involved in weights training, thus alleviating search instability. We demonstrate that under objectives of accuracy and potential, EMARS can balance exploitation and exploration, i.e., reducing search time and finding better models. Our searched models are small and perform better than prior works on three public COVID-19 3D CT datasets.
64.
Al-Sabri, Raeed; Gao, Jianliang; Chen, Jiamin; Oloulade, Babatounde Moctard; Lyu, Tengfei
Multi-View Graph Neural Architecture Search for Biomedical Entity and Relation Extraction Journal Article
In: IEEE/ACM Transactions on Computational Biology and Bioinformatics, pp. 1-13, 2022.
@article{9881878,
title = {Multi-View Graph Neural Architecture Search for Biomedical Entity and Relation Extraction},
author = {Raeed Al-Sabri and Jianliang Gao and Jiamin Chen and Babatounde Moctard Oloulade and Tengfei Lyu},
url = {https://ieeexplore.ieee.org/abstract/document/9881878},
doi = {10.1109/TCBB.2022.3205113},
year = {2022},
date = {2022-01-01},
urldate = {2022-01-01},
journal = {IEEE/ACM Transactions on Computational Biology and Bioinformatics},
pages = {1-13},
keywords = {},
pubstate = {published},
tppubtype = {article}
}