Contrastive learning of heart and lung sounds for label-efficient diagnosis

Pratham N Soni; Siyu Shi; Pranav R Sriram; Andrew Y Ng; Pranav Rajpurkar

doi:10.1016/j.patter.2021.100400

Contrastive learning of heart and lung sounds for label-efficient diagnosis

Patterns (N Y). 2021 Dec 7;3(1):100400. doi: 10.1016/j.patter.2021.100400. eCollection 2022 Jan 14.

Authors

Pratham N Soni¹, Siyu Shi², Pranav R Sriram¹, Andrew Y Ng¹, Pranav Rajpurkar³

Affiliations

¹ Department of Computer Science, Stanford University, Stanford, CA, USA.
² School of Medicine, Stanford University, Stanford, CA, USA.
³ Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA.

Abstract

Data labeling is often the limiting step in machine learning because it requires time from trained experts. To address the limitation on labeled data, contrastive learning, among other unsupervised learning methods, leverages unlabeled data to learn representations of data. Here, we propose a contrastive learning framework that utilizes metadata for selecting positive and negative pairs when training on unlabeled data. We demonstrate its application in the healthcare domain on heart and lung sound recordings. The increasing availability of heart and lung sound recordings due to adoption of digital stethoscopes lends itself as an opportunity to demonstrate the application of our contrastive learning method. Compared to contrastive learning with augmentations, the contrastive learning model leveraging metadata for pair selection utilizes clinical information associated with lung and heart sound recordings. This approach uses shared context of the recordings on the patient level using clinical information including age, sex, weight, location of sounds, etc. We show improvement in downstream tasks for diagnosing heart and lung sounds when leveraging patient-specific representations in selecting positive and negative pairs. This study paves the path for medical applications of contrastive learning that leverage clinical information. We have made our code available here: https://github.com/stanfordmlgroup/selfsupervised-lungandheartsounds.

Keywords: contrastive learning; heart sounds; lung sounds; medicine; self-supervised learning; unlabeled data.