MedNAS: Multiscale Training-Free Neural Architecture Search for Medical Image Analysis

Wang, Yan; Zhen, Liangli; Zhang, Jianwei; Li, Miqing; Zhang, Lei; Wang, Zizhou; Feng, Yangqin; Xue, Yu; Wang, Xiao; Chen, Zheng; Luo, Tao; Goh, Rick Mong; Liu, Yong

doi:10.1109/tevc.2024.3352641

MedNAS: Multiscale Training-Free Neural Architecture Search for Medical Image Analysis

Journal Article · Thu Jan 11 04:00:00 EST 2024 · IEEE Transactions on Evolutionary Computation

DOI:https://doi.org/10.1109/tevc.2024.3352641· OSTI ID:2439002

^[1]; ^[1]; ^[2]; ^[3]; ^[4]; Wang, Zizhou ^[1]; Feng, Yangqin ^[1]; ^[5]; ^[6]; ^[7]; ^[1]; ^[1]; Liu, Yong ^[1]

Agency for Science, Technology and Research (A*STAR) (Singapore). Institute of High Performance Computing
Chengdu University of Information Technology (China); Sichuan Univ., Chengdu (China)
Univ. of Birmingham (United Kingdom)
Sichuan Univ., Chengdu (China)
Nanjing Univ. of Information Science and Technology (China)
Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
Osaka Univ. (Japan)

Deep neural networks have demonstrated impressive results in medical image analysis, but designing suitable architectures for each specific task is expertise dependent and time consuming. Neural architecture search (NAS) offers an effective means of discovering architectures. It has been highly successful in numerous applications, particularly in natural image classification. Yet, medical images possess unique characteristics, such as small regions and a wide variety of lesion sizes, that differentiate them from natural images. Furthermore, most current NAS methods struggle with high computational costs, especially when dealing with high-resolution image datasets. In this article, we present a novel evolutionary NAS method called multiscale training-free neural architecture search (MSTF-NAS) to address these challenges. Specifically, to accommodate the broad range of lesion region sizes in disease diagnosis, we develop a new reduction cell search space that enables the search algorithm to explicitly identify the optimal scale combination for multiscale feature extraction. Further, to overcome the issue of high computational costs, we utilize training-free indicators as performance measures for candidate architectures, which allows us to search for the optimal architecture more efficiently. More specifically, by considering the capability and simplicity of various networks, we formulate a multiobjective optimization problem that involves two training-free indicators and model complexity for candidate architectures. Extensive experiments on a large medical image benchmark and a publicly available breast cancer detection dataset are conducted. The empirical results demonstrate that our MSTF-NAS outperforms both human-designed architectures and current state-of-the-art NAS algorithms on both datasets, indicating the effectiveness of our proposed method.

View Accepted Manuscript (DOE)

Research Organization:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: USDOE

Grant/Contract Number:: AC05-00OR22725

OSTI ID:: 2439002

Journal Information:: IEEE Transactions on Evolutionary Computation, Journal Name: IEEE Transactions on Evolutionary Computation Journal Issue: 3 Vol. 28; ISSN 1089-778X

Publisher:: IEEECopyright Statement

Country of Publication:: United States

Language:: English

References (25)

Neural Architecture Search Elsken, Thomas; Metzen, Jan Hendrik; Hutter, Frank Automated Machine Learning https://doi.org/10.1007/978-3-030-05318-5_3	book	January 2019
U-Net: Convolutional Networks for Biomedical Image Segmentation Ronneberger, Olaf; Fischer, Philipp; Brox, Thomas Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III https://doi.org/10.1007/978-3-319-24574-4_28	book	November 2015
NSGA-II algorithm for multi-objective generation expansion planning problem Murugan, P.; Kannan, S.; Baskar, S. Electric Power Systems Research, Vol. 79, Issue 4 https://doi.org/10.1016/j.epsr.2008.09.011	journal	April 2009
A curated mammography data set for use in computer-aided detection and diagnosis research Lee, Rebecca Sawyer; Gimenez, Francisco; Hoogi, Assaf Scientific Data, Vol. 4, Issue 1 https://doi.org/10.1038/sdata.2017.177	journal	December 2017
A fast and elitist multiobjective genetic algorithm: NSGA-II Deb, K.; Pratap, A.; Agarwal, S. IEEE Transactions on Evolutionary Computation, Vol. 6, Issue 2 https://doi.org/10.1109/4235.996017	journal	April 2002
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation Girshick, Ross; Donahue, Jeff; Darrell, Trevor 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2014.81	conference	June 2014
Attention to Scale: Scale-Aware Semantic Image Segmentation Chen, Liang-Chieh; Yang, Yi; Wang, Jiang 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2016.396	conference	June 2016
Deep Residual Learning for Image Recognition He, Kaiming; Zhang, Xiangyu; Ren, Shaoqing 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2016.90	conference	June 2016
You Only Look Once: Unified, Real-Time Object Detection Redmon, Joseph; Divvala, Santosh; Girshick, Ross 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2016.91	conference	June 2016
Densely Connected Convolutional Networks Huang, Gao; Liu, Zhuang; Maaten, Laurens van der 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2017.243	conference	July 2017
Squeeze-and-Excitation Networks Hu, Jie; Shen, Li; Sun, Gang 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2018.00745	conference	June 2018
FBNetV2: Differentiable Neural Architecture Search for Spatial and Channel Dimensions Wan, Alvin; Dai, Xiaoliang; Zhang, Peizhao 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR42600.2020.01298	conference	June 2020
Towards Improving the Consistency, Efficiency, and Flexibility of Differentiable Neural Architecture Search Yang, Yibo; You, Shan; Li, Hongyang 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR46437.2021.00660	conference	June 2021
Segmenter: Transformer for Semantic Segmentation Strudel, Robin; Garcia, Ricardo; Laptev, Ivan 2021 IEEE/CVF International Conference on Computer Vision (ICCV) https://doi.org/10.1109/ICCV48922.2021.00717	conference	October 2021
Fast, Accurate and Lightweight Super-Resolution with Neural Architecture Search Chu, Xiangxiang; Zhang, Bo; Ma, Hailong 2020 25th International Conference on Pattern Recognition (ICPR) https://doi.org/10.1109/ICPR48806.2021.9413080	conference	January 2021
Automatically Designing CNN Architectures Using the Genetic Algorithm for Image Classification Sun, Yanan; Xue, Bing; Zhang, Mengjie IEEE Transactions on Cybernetics, Vol. 50, Issue 9 https://doi.org/10.1109/TCYB.2020.2983860	journal	September 2020
BenchENAS: A Benchmarking Platform for Evolutionary Neural Architecture Search Xie, Xiangning; Liu, Yuqiao; Sun, Yanan IEEE Transactions on Evolutionary Computation, Vol. 26, Issue 6 https://doi.org/10.1109/TEVC.2022.3147526	journal	December 2022
FreeREA: Training-Free Evolution-based Architecture Search Cavagnero, Niccolo; Robbiano, Luca; Caputo, Barbara 2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) https://doi.org/10.1109/WACV56688.2023.00154	conference	January 2023
Revisiting Training-free NAS Metrics: An Efficient Training-based Method Yang, Taojiannan; Yang, Linjie; Jin, Xiaojie 2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) https://doi.org/10.1109/WACV56688.2023.00473	conference	January 2023
One-Shot Neural Architecture Search by Dynamically Pruning Supernet in Hierarchical Order Zhang, Jianwei; Li, Dong; Wang, Lituan International Journal of Neural Systems, Vol. 31, Issue 07 https://doi.org/10.1142/S0129065721500295	journal	June 2021
ImageNet classification with deep convolutional neural networks Krizhevsky, Alex; Sutskever, Ilya; Hinton, Geoffrey E. Communications of the ACM, Vol. 60, Issue 6 https://doi.org/10.1145/3065386	journal	May 2017
Auto-Keras: An Efficient Neural Architecture Search System Jin, Haifeng; Song, Qingquan; Hu, Xia Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining https://doi.org/10.1145/3292500.3330648	conference	July 2019
NSGA-Net Lu, Zhichao; Whalen, Ian; Boddeti, Vishnu Proceedings of the Genetic and Evolutionary Computation Conference https://doi.org/10.1145/3321707.3321729	conference	July 2019
MSU-Net: Multi-Scale U-Net for 2D Medical Image Segmentation Su, Run; Zhang, Deyun; Liu, Jinhuai Frontiers in Genetics, Vol. 12 https://doi.org/10.3389/fgene.2021.639930	journal	February 2021
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale Dosovitskiy, Alexey; Beyer, Lucas; Kolesnikov, Alexander arXiv https://doi.org/10.48550/arXiv.2010.11929	preprint	January 2020

Similar Records

Explainable Neural Architecture Search (XNAS)

Software · Thu Dec 09 19:00:00 EST 2021 · OSTI ID:code-70518

Evolutionary Architecture Search for Generative Adversarial Networks Based on Weight Sharing

Journal Article · Thu Nov 30 23:00:00 EST 2023 · IEEE Transactions on Evolutionary Computation · OSTI ID:2429864

Related Subjects

60 APPLIED LIFE SCIENCES
evolutionary optimization
feature extraction
lesions
medical diagnostic imaging
medical image classification
neural architecture search
optimization
training-free architecture search

MedNAS: Multiscale Training-Free Neural Architecture Search for Medical Image Analysis

Citation Formats

References (25)

Similar Records

Related Subjects