An Explainable Artificial Intelligence Text Classifier for Suicidality Prediction in Youth Crisis Text Line Users: Development and Validation Study
Neural networks using large language model-based transfer learning can accurately identify SI and ASE. The post hoc explainer model revealed language features associated with SI and ASE. Such models may potentially support clinical decision-making in suicide prevention services. Future research should explore multimodal input features and temporal aspects of suicide risk.
Authors
- Lars Kuchinke (ORCID: https://orcid.org/0000-0001-8248-1167)
- Gunther Meinlschmidt (ORCID: https://orcid.org/0000-0002-3488-193X)
- Julia Thomas (ORCID: https://orcid.org/0000-0002-2444-3389)
- Marcel Miché (ORCID: https://orcid.org/0000-0001-8838-1749)
- Richard Wundrack (ORCID: https://orcid.org/0000-0003-2121-0982)
- Antonia Lucht (ORCID: https://orcid.org/0000-0002-5106-0340)
- Jacob Segler (ORCID: https://orcid.org/0000-0001-6694-5507)
- Roselind Lieb (ORCID: https://orcid.org/0000-0002-2039-2262)
Institutions
- Trier University of Applied Sciences (DE)
- University of Basel (CH)
- University Hospital of Basel (CH)
- International Psychoanalytic University Berlin (DE)
- Climate Analytics (DE)
- University Hospital Ulm (DE)
- Universität Trier (DE)
Publication Details
- Journal
- JMIR Public Health and Surveillance
- Published
- 2024-11-07
- DOI
- https://doi.org/10.2196/63809
- Citations
- 11
- Primary Topic
- Suicide and Self-Harm Studies
- Type
- article
- Field-Weighted Citation Impact
- 7.03