HeartMAE: Advancing Cardiac MRI Analysis through Optical Flow Guided Masked Autoencoding

Vladislav Kim, Lisa Schneider, Soodeh Kalaie, Declan O’Regan, Christian Bender
Proceedings of the 4th Machine Learning for Health Symposium, PMLR 259:594-609, 2025.

Abstract

Cardiac MRI is a powerful diagnostic tool, but traditional analysis relies on complex segmentation-based workflows that may provide only a partial picture of cardiovascular health. To address these limitations, we introduce HeartMAE, a novel framework that uses masked autoencoding (MAE) to learn features directly from cardiac MRIs, without any labels. By incorporating optical flow during training, HeartMAE is guided towards cardiac regions, which significantly improves its downstream performance. A multitask model, built on a shared HeartMAE embedding layer, accurately predicts key cardiac health indicators, extracardiac features and major cardiovascular conditions. Moreover, HeartMAE features may be used as embeddings for clustering to enable patient stratification. Requiring only MRI data, HeartMAE is highly scalable and adaptable to larger datasets, paving the way for foundation models in cardiac imaging.

Cite this Paper


BibTeX
@InProceedings{pmlr-v259-kim25a, title = {HeartMAE: Advancing Cardiac MRI Analysis through Optical Flow Guided Masked Autoencoding}, author = {Kim, Vladislav and Schneider, Lisa and Kalaie, Soodeh and O'Regan, Declan and Bender, Christian}, booktitle = {Proceedings of the 4th Machine Learning for Health Symposium}, pages = {594--609}, year = {2025}, editor = {Hegselmann, Stefan and Zhou, Helen and Healey, Elizabeth and Chang, Trenton and Ellington, Caleb and Mhasawade, Vishwali and Tonekaboni, Sana and Argaw, Peniel and Zhang, Haoran}, volume = {259}, series = {Proceedings of Machine Learning Research}, month = {15--16 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v259/main/assets/kim25a/kim25a.pdf}, url = {https://proceedings.mlr.press/v259/kim25a.html}, abstract = {Cardiac MRI is a powerful diagnostic tool, but traditional analysis relies on complex segmentation-based workflows that may provide only a partial picture of cardiovascular health. To address these limitations, we introduce HeartMAE, a novel framework that uses masked autoencoding (MAE) to learn features directly from cardiac MRIs, without any labels. By incorporating optical flow during training, HeartMAE is guided towards cardiac regions, which significantly improves its downstream performance. A multitask model, built on a shared HeartMAE embedding layer, accurately predicts key cardiac health indicators, extracardiac features and major cardiovascular conditions. Moreover, HeartMAE features may be used as embeddings for clustering to enable patient stratification. Requiring only MRI data, HeartMAE is highly scalable and adaptable to larger datasets, paving the way for foundation models in cardiac imaging.} }
Endnote
%0 Conference Paper %T HeartMAE: Advancing Cardiac MRI Analysis through Optical Flow Guided Masked Autoencoding %A Vladislav Kim %A Lisa Schneider %A Soodeh Kalaie %A Declan O’Regan %A Christian Bender %B Proceedings of the 4th Machine Learning for Health Symposium %C Proceedings of Machine Learning Research %D 2025 %E Stefan Hegselmann %E Helen Zhou %E Elizabeth Healey %E Trenton Chang %E Caleb Ellington %E Vishwali Mhasawade %E Sana Tonekaboni %E Peniel Argaw %E Haoran Zhang %F pmlr-v259-kim25a %I PMLR %P 594--609 %U https://proceedings.mlr.press/v259/kim25a.html %V 259 %X Cardiac MRI is a powerful diagnostic tool, but traditional analysis relies on complex segmentation-based workflows that may provide only a partial picture of cardiovascular health. To address these limitations, we introduce HeartMAE, a novel framework that uses masked autoencoding (MAE) to learn features directly from cardiac MRIs, without any labels. By incorporating optical flow during training, HeartMAE is guided towards cardiac regions, which significantly improves its downstream performance. A multitask model, built on a shared HeartMAE embedding layer, accurately predicts key cardiac health indicators, extracardiac features and major cardiovascular conditions. Moreover, HeartMAE features may be used as embeddings for clustering to enable patient stratification. Requiring only MRI data, HeartMAE is highly scalable and adaptable to larger datasets, paving the way for foundation models in cardiac imaging.
APA
Kim, V., Schneider, L., Kalaie, S., O’Regan, D. & Bender, C.. (2025). HeartMAE: Advancing Cardiac MRI Analysis through Optical Flow Guided Masked Autoencoding. Proceedings of the 4th Machine Learning for Health Symposium, in Proceedings of Machine Learning Research 259:594-609 Available from https://proceedings.mlr.press/v259/kim25a.html.

Related Material