MedFuse: Multi-modal fusion with clinical time-series data and chest X-ray images

Nasir Hayat, Krzysztof J. Geras, Farah E. Shamout
Proceedings of the 7th Machine Learning for Healthcare Conference, PMLR 182:479-503, 2022.

Abstract

Multi-modal fusion approaches aim to integrate information from different data sources. Unlike natural datasets, such as in audio-visual applications, where samples consist of “paired” modalities, data in healthcare is often collected asynchronously. Hence, requiring the presence of all modalities for a given sample is not realistic for clinical tasks and significantly limits the size of the dataset during training. In this paper, we propose MedFuse, a conceptually simple yet promising LSTM-based fusion module that can accommodate uni-modal as well as multi-modal input. We evaluate the fusion method and introduce new benchmark results for in-hospital mortality prediction and phenotype classification, using clinical time-series data in the MIMIC-IV dataset and corresponding chest X-ray images in MIMIC-CXR. Compared to more complex multi-modal fusion strategies, MedFuse provides a performance improvement by a large margin on the fully paired test set. It also remains robust across the partially paired test set containing samples with missing chest X-ray images. We release our code for reproducibility and to enable the evaluation of competing models in the future.

Cite this Paper

BibTeX
@InProceedings{pmlr-v182-hayat22a, title = {MedFuse: Multi-modal fusion with clinical time-series data and chest X-ray images}, author = {Hayat, Nasir and Geras, Krzysztof J. and Shamout, Farah E.}, booktitle = {Proceedings of the 7th Machine Learning for Healthcare Conference}, pages = {479--503}, 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/hayat22a/hayat22a.pdf}, url = {https://proceedings.mlr.press/v182/hayat22a.html}, abstract = {Multi-modal fusion approaches aim to integrate information from different data sources. Unlike natural datasets, such as in audio-visual applications, where samples consist of “paired” modalities, data in healthcare is often collected asynchronously. Hence, requiring the presence of all modalities for a given sample is not realistic for clinical tasks and significantly limits the size of the dataset during training. In this paper, we propose MedFuse, a conceptually simple yet promising LSTM-based fusion module that can accommodate uni-modal as well as multi-modal input. We evaluate the fusion method and introduce new benchmark results for in-hospital mortality prediction and phenotype classification, using clinical time-series data in the MIMIC-IV dataset and corresponding chest X-ray images in MIMIC-CXR. Compared to more complex multi-modal fusion strategies, MedFuse provides a performance improvement by a large margin on the fully paired test set. It also remains robust across the partially paired test set containing samples with missing chest X-ray images. We release our code for reproducibility and to enable the evaluation of competing models in the future.} }
Endnote
%0 Conference Paper %T MedFuse: Multi-modal fusion with clinical time-series data and chest X-ray images %A Nasir Hayat %A Krzysztof J. Geras %A Farah E. Shamout %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-hayat22a %I PMLR %P 479--503 %U https://proceedings.mlr.press/v182/hayat22a.html %V 182 %X Multi-modal fusion approaches aim to integrate information from different data sources. Unlike natural datasets, such as in audio-visual applications, where samples consist of “paired” modalities, data in healthcare is often collected asynchronously. Hence, requiring the presence of all modalities for a given sample is not realistic for clinical tasks and significantly limits the size of the dataset during training. In this paper, we propose MedFuse, a conceptually simple yet promising LSTM-based fusion module that can accommodate uni-modal as well as multi-modal input. We evaluate the fusion method and introduce new benchmark results for in-hospital mortality prediction and phenotype classification, using clinical time-series data in the MIMIC-IV dataset and corresponding chest X-ray images in MIMIC-CXR. Compared to more complex multi-modal fusion strategies, MedFuse provides a performance improvement by a large margin on the fully paired test set. It also remains robust across the partially paired test set containing samples with missing chest X-ray images. We release our code for reproducibility and to enable the evaluation of competing models in the future.
APA
Hayat, N., Geras, K.J. & Shamout, F.E.. (2022). MedFuse: Multi-modal fusion with clinical time-series data and chest X-ray images. Proceedings of the 7th Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 182:479-503 Available from https://proceedings.mlr.press/v182/hayat22a.html.

Related Material