Contrastive learning improves critical event prediction in COVID-19 patients

Tingyi Wanyan, Hossein Honarvar, Suraj K. Jaladanki, Chengxi Zang, Nidhi Naik, Sulaiman Somani, Jessica K. De Freitas, Ishan Paranjpe, Akhil Vaid, Jing Zhang, Riccardo Miotto, Zhangyang Wang, Girish N. Nadkarni, Marinka Zitnik, Ariful Azad, Fei Wang, Ying Ding, Benjamin S. Glicksberg

Research output: Contribution to journalArticlepeer-review

18 Scopus citations

Abstract

Deep learning (DL) models typically require large-scale, balanced training data to be robust, generalizable, and effective in the context of healthcare. This has been a major issue for developing DL models for the coronavirus disease 2019 (COVID-19) pandemic, where data are highly class imbalanced. Conventional approaches in DL use cross-entropy loss (CEL), which often suffers from poor margin classification. We show that contrastive loss (CL) improves the performance of CEL, especially in imbalanced electronic health records (EHR) data for COVID-19 analyses. We use a diverse EHR dataset to predict three outcomes: mortality, intubation, and intensive care unit (ICU) transfer in hospitalized COVID-19 patients over multiple time windows. To compare the performance of CEL and CL, models are tested on the full dataset and a restricted dataset. CL models consistently outperform CEL models, with differences ranging from 0.04 to 0.15 for area under the precision and recall curve (AUPRC) and 0.05 to 0.1 for area under the receiver-operating characteristic curve (AUROC).

Original languageEnglish
Article number100389
JournalPatterns
Volume2
Issue number12
DOIs
StatePublished - 10 Dec 2021

Keywords

  • COVID-19
  • DSML 3: Development/Pre-production: Data science output has been rolled out/validated across multiple domains/problems
  • contrastive loss
  • deep learning
  • electronic health records
  • machine learning
  • recurrent neural network

Fingerprint

Dive into the research topics of 'Contrastive learning improves critical event prediction in COVID-19 patients'. Together they form a unique fingerprint.

Cite this