A Reproducible Model Based on Clinical Text for Predicting Suicidal Behavior.
| Autor: | Obeid JS; Medical University of South Carolina, Charleston, SC, USA., Tsalatsanis A; University of South Florida, Tampa, FL, USA., Chaphalkar C; University of South Florida, Tampa, FL, USA., Robinson S; University of South Florida, Tampa, FL, USA., Klein S; University of South Florida, Tampa, FL, USA., Cool S; University of South Florida, Tampa, FL, USA., Szwast E; Medical University of South Carolina, Charleston, SC, USA., Bunnell BE; University of South Florida, Tampa, FL, USA. |
|---|---|
| Jazyk: | angličtina |
| Zdroj: | Studies in health technology and informatics [Stud Health Technol Inform] 2024 Jan 25; Vol. 310, pp. 1486-1487. |
| DOI: | 10.3233/SHTI231257 |
| Abstrakt: | Suicide risk models are critical for prioritizing patients for intervention. We demonstrate a reproducible approach for training text classifiers to identify patients at risk. The models were effective in phenotyping suicidal behavior (F1=.94) and moderately effective in predicting future events (F1=.63). |
| Databáze: | MEDLINE |
| Externí odkaz: |
|