DAFTED: Decoupled Asymmetric Fusion of Tabular and Echocardiographic Data for Cardiac Hypertension Diagnosis

Jérémie Stym-Popper, Nathan Painchaud, Pierre-Yves Courand, Clément Rambour, Nicolas Thome, Olivier Bernard
Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, PMLR 301:1465-1482, 2026.

Abstract

Multimodal data fusion has emerged as a key approach in recent years for enhancing diagnosis and prognosis in many medical applications. With the advent of transformer-based methods, it is now possible to combine information from different modalities that provide complementary insights. However, most existing methods rely on symmetric fusion schemes, assuming equal importance for information carried by each modality—a strong assumption that may not always hold true. In this study, we propose an alternative fusion strategy based on an asymmetric scheme. Starting with a primary modality that offers the most critical information, we integrate secondary modality contributions by disentangling shared and modality-specific information. The proposed model was validated on a dataset of 239 patients for characterizing hypertension severity by fusing time series automatically extracted from echocardiographic image sequences and tabular data from patient records. Results show that our approach outperforms existing unimodal and multimodal approaches, achieving an AUC score over 90% - a crucial benchmark for clinical use.

Cite this Paper

BibTeX
@InProceedings{pmlr-v301-stym-popper26a, title = {DAFTED: Decoupled Asymmetric Fusion of Tabular and Echocardiographic Data for Cardiac Hypertension Diagnosis}, author = {Stym-Popper, J\'er\'emie and Painchaud, Nathan and Courand, Pierre-Yves and Rambour, Cl\'ement and Thome, Nicolas and Bernard, Olivier}, booktitle = {Proceedings of The 8th International Conference on Medical Imaging with Deep Learning}, pages = {1465--1482}, year = {2026}, editor = {Tasdizen, Tolga and Elhabian, Shireen and Summers, Ronald and Chen, Chen and Koch, Lisa and Zhuang, Yan}, volume = {301}, series = {Proceedings of Machine Learning Research}, month = {09--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v301/main/assets/stym-popper26a/stym-popper26a.pdf}, url = {https://proceedings.mlr.press/v301/stym-popper26a.html}, abstract = {Multimodal data fusion has emerged as a key approach in recent years for enhancing diagnosis and prognosis in many medical applications. With the advent of transformer-based methods, it is now possible to combine information from different modalities that provide complementary insights. However, most existing methods rely on symmetric fusion schemes, assuming equal importance for information carried by each modality—a strong assumption that may not always hold true. In this study, we propose an alternative fusion strategy based on an asymmetric scheme. Starting with a primary modality that offers the most critical information, we integrate secondary modality contributions by disentangling shared and modality-specific information. The proposed model was validated on a dataset of 239 patients for characterizing hypertension severity by fusing time series automatically extracted from echocardiographic image sequences and tabular data from patient records. Results show that our approach outperforms existing unimodal and multimodal approaches, achieving an AUC score over 90% - a crucial benchmark for clinical use.} }
Endnote
%0 Conference Paper %T DAFTED: Decoupled Asymmetric Fusion of Tabular and Echocardiographic Data for Cardiac Hypertension Diagnosis %A Jérémie Stym-Popper %A Nathan Painchaud %A Pierre-Yves Courand %A Clément Rambour %A Nicolas Thome %A Olivier Bernard %B Proceedings of The 8th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Tolga Tasdizen %E Shireen Elhabian %E Ronald Summers %E Chen Chen %E Lisa Koch %E Yan Zhuang %F pmlr-v301-stym-popper26a %I PMLR %P 1465--1482 %U https://proceedings.mlr.press/v301/stym-popper26a.html %V 301 %X Multimodal data fusion has emerged as a key approach in recent years for enhancing diagnosis and prognosis in many medical applications. With the advent of transformer-based methods, it is now possible to combine information from different modalities that provide complementary insights. However, most existing methods rely on symmetric fusion schemes, assuming equal importance for information carried by each modality—a strong assumption that may not always hold true. In this study, we propose an alternative fusion strategy based on an asymmetric scheme. Starting with a primary modality that offers the most critical information, we integrate secondary modality contributions by disentangling shared and modality-specific information. The proposed model was validated on a dataset of 239 patients for characterizing hypertension severity by fusing time series automatically extracted from echocardiographic image sequences and tabular data from patient records. Results show that our approach outperforms existing unimodal and multimodal approaches, achieving an AUC score over 90% - a crucial benchmark for clinical use.
APA
Stym-Popper, J., Painchaud, N., Courand, P., Rambour, C., Thome, N. & Bernard, O.. (2026). DAFTED: Decoupled Asymmetric Fusion of Tabular and Echocardiographic Data for Cardiac Hypertension Diagnosis. Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 301:1465-1482 Available from https://proceedings.mlr.press/v301/stym-popper26a.html.

Related Material