Leveraging Deep Representations of Radiology Reports in Survival Analysis for Predicting Heart Failure Patient Mortality
| Autor: | Hyun Gi Lee, Ashley Beecy, Subhi J. Al'Aref, Yifan Peng, Evan Sholle |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
0301 basic medicine Computer science Machine learning computer.software_genre Article 03 medical and health sciences 0302 clinical medicine Covariate medicine 030212 general & internal medicine Survival analysis Computer Science - Computation and Language business.industry Proportional hazards model Baseline model Patient survival medicine.disease 3. Good health Textual information 030104 developmental biology Heart failure Artificial intelligence Hidden layer business Computation and Language (cs.CL) computer |
| Zdroj: | NAACL-HLT Proc Conf |
| DOI: | 10.18653/v1/2021.naacl-main.358 |
| Popis: | 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. Comment: NAACL-HLT 2021, Short Paper |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|