Addressing The False Negative Problem of Deep Learning MRI Reconstruction Models by Adversarial Attacks and Robust Training

Kaiyang Cheng, Francesco Calivá, Rutwik Shah, Misung Han, Sharmila Majumdar, Valentina Pedoia
; Proceedings of the Third Conference on Medical Imaging with Deep Learning, PMLR 121:121-135, 2020.

Abstract

Deep learning models have been shown to be successful in accelerating MRI reconstruction, over traditional methods. However, it has been observed that these methods tend to miss rare small features, such as meniscal tears, subchondral osteophyte, etc. in musculoskeletal applications. This is a concerning finding as these small and rare features are the particularly relevant in clinical diagnostic settings. Additionally, such potentially dangerous loss of details in the reconstructed images are not reflected by global image fidelity metrics such as mean-square error (MSE) and structural similarity metric (SSIM). In this work, we propose a framework to find the worst-case false negatives by adversarially attacking the trained models and improve the models’ability to reconstruct the small features by robust training.

Cite this Paper


BibTeX
@InProceedings{pmlr-v121-cheng20a, title = {Addressing The False Negative Problem of Deep Learning MRI Reconstruction Models by Adversarial Attacks and Robust Training}, author = {Cheng, Kaiyang and Caliv\'a, Francesco and Shah, Rutwik and Han, Misung and Majumdar, Sharmila and Pedoia, Valentina}, pages = {121--135}, year = {2020}, editor = {Tal Arbel and Ismail Ben Ayed and Marleen de Bruijne and Maxime Descoteaux and Herve Lombaert and Christopher Pal}, volume = {121}, series = {Proceedings of Machine Learning Research}, address = {Montreal, QC, Canada}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v121/cheng20a/cheng20a.pdf}, url = {http://proceedings.mlr.press/v121/cheng20a.html}, abstract = {Deep learning models have been shown to be successful in accelerating MRI reconstruction, over traditional methods. However, it has been observed that these methods tend to miss rare small features, such as meniscal tears, subchondral osteophyte, etc. in musculoskeletal applications. This is a concerning finding as these small and rare features are the particularly relevant in clinical diagnostic settings. Additionally, such potentially dangerous loss of details in the reconstructed images are not reflected by global image fidelity metrics such as mean-square error (MSE) and structural similarity metric (SSIM). In this work, we propose a framework to find the worst-case false negatives by adversarially attacking the trained models and improve the models’ability to reconstruct the small features by robust training.} }
Endnote
%0 Conference Paper %T Addressing The False Negative Problem of Deep Learning MRI Reconstruction Models by Adversarial Attacks and Robust Training %A Kaiyang Cheng %A Francesco Calivá %A Rutwik Shah %A Misung Han %A Sharmila Majumdar %A Valentina Pedoia %B Proceedings of the Third Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2020 %E Tal Arbel %E Ismail Ben Ayed %E Marleen de Bruijne %E Maxime Descoteaux %E Herve Lombaert %E Christopher Pal %F pmlr-v121-cheng20a %I PMLR %J Proceedings of Machine Learning Research %P 121--135 %U http://proceedings.mlr.press %V 121 %W PMLR %X Deep learning models have been shown to be successful in accelerating MRI reconstruction, over traditional methods. However, it has been observed that these methods tend to miss rare small features, such as meniscal tears, subchondral osteophyte, etc. in musculoskeletal applications. This is a concerning finding as these small and rare features are the particularly relevant in clinical diagnostic settings. Additionally, such potentially dangerous loss of details in the reconstructed images are not reflected by global image fidelity metrics such as mean-square error (MSE) and structural similarity metric (SSIM). In this work, we propose a framework to find the worst-case false negatives by adversarially attacking the trained models and improve the models’ability to reconstruct the small features by robust training.
APA
Cheng, K., Calivá, F., Shah, R., Han, M., Majumdar, S. & Pedoia, V.. (2020). Addressing The False Negative Problem of Deep Learning MRI Reconstruction Models by Adversarial Attacks and Robust Training. Proceedings of the Third Conference on Medical Imaging with Deep Learning, in PMLR 121:121-135

Related Material