Work

Leveraging Deep Representations of Radiology Reports in Survival Analysis for Predicting Heart Failure Patient Mortality

Year: 2021

Type: conference-paper

Abstract: Utilizing clinical texts in survival analysis is difficult because they are largely unstructured. Current automatic extraction models fail to capture textual information comprehensively since their labels are limited in scope. Furthermore, they typically require a large amount of data and high-quality expert annotations for training. In this work, we present a novel method of using BERT-based hidden layer representations of clinical texts as covariates for proportional hazards models to predict patient survival outcomes. We show that hidden layers yield notably more accurate predictions than predefined features, outperforming the previous baseline model by 5.7% on average across C-index and time-dependent AUC. We make our work publicly available at https://github.com/bionlplab/heart_failure_mortality.

Source: Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)

DOI: 10.18653/v1/2021.naacl-main.358

Authors: Hyun Gi Lee, Evan Sholle, Ashley Beecy, Subhi Al’Aref, Yifan Peng

Institutions: Weill Cornell Medicine, University of Arkansas Medical Center, University of Arkansas for Medical Sciences

Language: English

Topic: Machine Learning in Healthcare

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDG: Good health and well-being

Open Access status: hybrid