Deep Metric Learning for the Hemodynamics Inference with Electrocardiogram Signals

Hyewon Jeong, Collin M. Stultz, Marzyeh Ghassemi
Proceedings of the 8th Machine Learning for Healthcare Conference, PMLR 219:321-342, 2023.

Abstract

Heart failure is a debilitating condition that affects millions of people worldwide and has a significant impact on their quality of life and mortality rates. An objective assessment of cardiac pressures remains an important method for the diagnosis and treatment prognostication for patients with heart failure. Although cardiac catheterization is the gold standard for estimating central hemodynamic pressures, it is an invasive procedure that carries inherent risks, making it a potentially dangerous procedure for some patients. Approaches that leverage non-invasive signals – such as the electrocardiogram (ECG) – have the promise to make the routine estimation of cardiac pressures feasible in both inpatient and outpatient settings. Prior models trained to estimate intracardiac pressures (e.g., mean pulmonary capillary wedge pressure (mPCWP)) in a supervised fashion have shown good discriminatory ability but have been limited to the labeled dataset from the heart failure cohort. Furthermore, obtaining large datasets for diverse patient cohorts with intracardiac pressure labels is challenging due to the invasive nature of the procedure. To address these issues and build a robust representation, we apply traditional deep metric learning (DML) and propose a novel self-supervised DML with distance-based mining that improves the performance of a model with limited labels. We use a dataset that contains over 5.4 million ECGs without concomitant central pressure labels to pre-train a self-supervised DML model which showed improved classification of elevated mPCWP compared to self-supervised contrastive baselines. Additionally, the supervised DML model that uses ECGs with access to 8,172 mPCWP labels demonstrated significantly better performance on the mPCWP regression task compared to the supervised baseline. Moreover, our data suggest that DML yields models that are performant across patient subgroups, even when some patient subgroups are under-represented in the dataset.

Cite this Paper

BibTeX
@InProceedings{pmlr-v219-jeong23a, title = {Deep Metric Learning for the Hemodynamics Inference with Electrocardiogram Signals}, author = {Jeong, Hyewon and Stultz, Collin M. and Ghassemi, Marzyeh}, booktitle = {Proceedings of the 8th Machine Learning for Healthcare Conference}, pages = {321--342}, year = {2023}, editor = {Deshpande, Kaivalya and Fiterau, Madalina and Joshi, Shalmali and Lipton, Zachary and Ranganath, Rajesh and Urteaga, Iñigo and Yeung, Serene}, volume = {219}, series = {Proceedings of Machine Learning Research}, month = {11--12 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v219/jeong23a/jeong23a.pdf}, url = {https://proceedings.mlr.press/v219/jeong23a.html}, abstract = {Heart failure is a debilitating condition that affects millions of people worldwide and has a significant impact on their quality of life and mortality rates. An objective assessment of cardiac pressures remains an important method for the diagnosis and treatment prognostication for patients with heart failure. Although cardiac catheterization is the gold standard for estimating central hemodynamic pressures, it is an invasive procedure that carries inherent risks, making it a potentially dangerous procedure for some patients. Approaches that leverage non-invasive signals – such as the electrocardiogram (ECG) – have the promise to make the routine estimation of cardiac pressures feasible in both inpatient and outpatient settings. Prior models trained to estimate intracardiac pressures (e.g., mean pulmonary capillary wedge pressure (mPCWP)) in a supervised fashion have shown good discriminatory ability but have been limited to the labeled dataset from the heart failure cohort. Furthermore, obtaining large datasets for diverse patient cohorts with intracardiac pressure labels is challenging due to the invasive nature of the procedure. To address these issues and build a robust representation, we apply traditional deep metric learning (DML) and propose a novel self-supervised DML with distance-based mining that improves the performance of a model with limited labels. We use a dataset that contains over 5.4 million ECGs without concomitant central pressure labels to pre-train a self-supervised DML model which showed improved classification of elevated mPCWP compared to self-supervised contrastive baselines. Additionally, the supervised DML model that uses ECGs with access to 8,172 mPCWP labels demonstrated significantly better performance on the mPCWP regression task compared to the supervised baseline. Moreover, our data suggest that DML yields models that are performant across patient subgroups, even when some patient subgroups are under-represented in the dataset.} }
Endnote
%0 Conference Paper %T Deep Metric Learning for the Hemodynamics Inference with Electrocardiogram Signals %A Hyewon Jeong %A Collin M. Stultz %A Marzyeh Ghassemi %B Proceedings of the 8th Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2023 %E Kaivalya Deshpande %E Madalina Fiterau %E Shalmali Joshi %E Zachary Lipton %E Rajesh Ranganath %E Iñigo Urteaga %E Serene Yeung %F pmlr-v219-jeong23a %I PMLR %P 321--342 %U https://proceedings.mlr.press/v219/jeong23a.html %V 219 %X Heart failure is a debilitating condition that affects millions of people worldwide and has a significant impact on their quality of life and mortality rates. An objective assessment of cardiac pressures remains an important method for the diagnosis and treatment prognostication for patients with heart failure. Although cardiac catheterization is the gold standard for estimating central hemodynamic pressures, it is an invasive procedure that carries inherent risks, making it a potentially dangerous procedure for some patients. Approaches that leverage non-invasive signals – such as the electrocardiogram (ECG) – have the promise to make the routine estimation of cardiac pressures feasible in both inpatient and outpatient settings. Prior models trained to estimate intracardiac pressures (e.g., mean pulmonary capillary wedge pressure (mPCWP)) in a supervised fashion have shown good discriminatory ability but have been limited to the labeled dataset from the heart failure cohort. Furthermore, obtaining large datasets for diverse patient cohorts with intracardiac pressure labels is challenging due to the invasive nature of the procedure. To address these issues and build a robust representation, we apply traditional deep metric learning (DML) and propose a novel self-supervised DML with distance-based mining that improves the performance of a model with limited labels. We use a dataset that contains over 5.4 million ECGs without concomitant central pressure labels to pre-train a self-supervised DML model which showed improved classification of elevated mPCWP compared to self-supervised contrastive baselines. Additionally, the supervised DML model that uses ECGs with access to 8,172 mPCWP labels demonstrated significantly better performance on the mPCWP regression task compared to the supervised baseline. Moreover, our data suggest that DML yields models that are performant across patient subgroups, even when some patient subgroups are under-represented in the dataset.
APA
Jeong, H., Stultz, C.M. & Ghassemi, M.. (2023). Deep Metric Learning for the Hemodynamics Inference with Electrocardiogram Signals. Proceedings of the 8th Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 219:321-342 Available from https://proceedings.mlr.press/v219/jeong23a.html.

Related Material