Anomaly Detection in Echocardiograms with Dynamic Variational Trajectory Models

Alain Ryser, Laura Manduchi, Fabian Laumer, Holger Michel, Sven Wellmann, Julia E. Vogt
Proceedings of the 7th Machine Learning for Healthcare Conference, PMLR 182:425-458, 2022.

Abstract

We propose a novel anomaly detection method for echocardiogram videos. The introduced method takes advantage of the periodic nature of the heart cycle to learn three variants of a variational latent trajectory model (TVAE). While the first two variants (TVAE-C and TVAE-R) model strict periodic movements of the heart, the third (TVAE-S) is more general and allows shifts in the spatial representation throughout the video. All models are trained on the healthy samples of a novel in-house dataset of infant echocardiogram videos consisting of multiple chamber views to learn a normative prior of the healthy population. During inference, maximum a posteriori (MAP) based anomaly detection is performed to detect out-of-distribution samples in our dataset. The proposed method reliably identifies severe congenital heart defects, such as Ebstein’s Anomaly or Shone-complex. Moreover, it achieves superior performance over MAP-based anomaly detection with standard variational autoencoders when detecting pulmonary hypertension and right ventricular dilation. Finally, we demonstrate that the proposed method enables interpretable explanations of its output through heatmaps highlighting the regions corresponding to anomalous heart structures.

Cite this Paper

BibTeX
@InProceedings{pmlr-v182-ryser22a, title = {Anomaly Detection in Echocardiograms with Dynamic Variational Trajectory Models}, author = {Ryser, Alain and Manduchi, Laura and Laumer, Fabian and Michel, Holger and Wellmann, Sven and Vogt, Julia E.}, booktitle = {Proceedings of the 7th Machine Learning for Healthcare Conference}, pages = {425--458}, year = {2022}, editor = {Lipton, Zachary and Ranganath, Rajesh and Sendak, Mark and Sjoding, Michael and Yeung, Serena}, volume = {182}, series = {Proceedings of Machine Learning Research}, month = {05--06 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v182/ryser22a/ryser22a.pdf}, url = {https://proceedings.mlr.press/v182/ryser22a.html}, abstract = {We propose a novel anomaly detection method for echocardiogram videos. The introduced method takes advantage of the periodic nature of the heart cycle to learn three variants of a variational latent trajectory model (TVAE). While the first two variants (TVAE-C and TVAE-R) model strict periodic movements of the heart, the third (TVAE-S) is more general and allows shifts in the spatial representation throughout the video. All models are trained on the healthy samples of a novel in-house dataset of infant echocardiogram videos consisting of multiple chamber views to learn a normative prior of the healthy population. During inference, maximum a posteriori (MAP) based anomaly detection is performed to detect out-of-distribution samples in our dataset. The proposed method reliably identifies severe congenital heart defects, such as Ebstein’s Anomaly or Shone-complex. Moreover, it achieves superior performance over MAP-based anomaly detection with standard variational autoencoders when detecting pulmonary hypertension and right ventricular dilation. Finally, we demonstrate that the proposed method enables interpretable explanations of its output through heatmaps highlighting the regions corresponding to anomalous heart structures.} }
Endnote
%0 Conference Paper %T Anomaly Detection in Echocardiograms with Dynamic Variational Trajectory Models %A Alain Ryser %A Laura Manduchi %A Fabian Laumer %A Holger Michel %A Sven Wellmann %A Julia E. Vogt %B Proceedings of the 7th Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2022 %E Zachary Lipton %E Rajesh Ranganath %E Mark Sendak %E Michael Sjoding %E Serena Yeung %F pmlr-v182-ryser22a %I PMLR %P 425--458 %U https://proceedings.mlr.press/v182/ryser22a.html %V 182 %X We propose a novel anomaly detection method for echocardiogram videos. The introduced method takes advantage of the periodic nature of the heart cycle to learn three variants of a variational latent trajectory model (TVAE). While the first two variants (TVAE-C and TVAE-R) model strict periodic movements of the heart, the third (TVAE-S) is more general and allows shifts in the spatial representation throughout the video. All models are trained on the healthy samples of a novel in-house dataset of infant echocardiogram videos consisting of multiple chamber views to learn a normative prior of the healthy population. During inference, maximum a posteriori (MAP) based anomaly detection is performed to detect out-of-distribution samples in our dataset. The proposed method reliably identifies severe congenital heart defects, such as Ebstein’s Anomaly or Shone-complex. Moreover, it achieves superior performance over MAP-based anomaly detection with standard variational autoencoders when detecting pulmonary hypertension and right ventricular dilation. Finally, we demonstrate that the proposed method enables interpretable explanations of its output through heatmaps highlighting the regions corresponding to anomalous heart structures.
APA
Ryser, A., Manduchi, L., Laumer, F., Michel, H., Wellmann, S. & Vogt, J.E.. (2022). Anomaly Detection in Echocardiograms with Dynamic Variational Trajectory Models. Proceedings of the 7th Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 182:425-458 Available from https://proceedings.mlr.press/v182/ryser22a.html.

Related Material