Automated Symptom Identification from Clinical Interview Transcripts in Diagnosed Schizophrenia: A Multi-Agent LLM Framework...36th Medical Informatics Europe (MIE) Conference, May 25-28, 2026, Genoa, Italy.
We present a multi-agent LLM framework for automated evaluation of symptom-criteria consistency in diagnosed schizophrenia cases. Our system decomposes evaluation into seven specialized agents with RAG-augmented DSM-5/ICD-11 knowledge: symptom extraction, standard matching, differential diagnosis, t...
| Publicado en: | Studies in Health Technology & Informatics Vol. 336; pp. 974 - 979 |
|---|---|
| Autores principales: | , |
| Formato: | equations & formulas proceedings research tables/charts Journal Article |
| Publicado: |
Sage Publications Inc.
2026
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=194018961&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 194018961 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 09269630 U1V jtl: Studies in Health Technology & Informatics issn: 09269630 maglogo: N pubinfo: dt: 2026 vid: 336 pid: 344 pub: Sage Publications Inc. place: Thousand Oaks, California artinfo: ui: 194018961 194018961 194018961 10.3233/SHTI260324 194018961 ppf: 974 ppct: 5 formats: tig: atl: Automated Symptom Identification from Clinical Interview Transcripts in Diagnosed Schizophrenia: A Multi-Agent LLM Framework...36th Medical Informatics Europe (MIE) Conference, May 25-28, 2026, Genoa, Italy. aug: au: ZHANG, Puzhen MAO, Jingzhi affil: Technical University of Munich, Munich, Germany. sug: subj: Electronic Health Records Schizophrenia Diagnosis Natural Language Processing Utilization Congresses and Conferences Italy Italy Human Conceptual Framework Interviews Confidence Intervals Diagnosis, Differential Voting Arbitration Descriptive Statistics ab: We present a multi-agent LLM framework for automated evaluation of symptom-criteria consistency in diagnosed schizophrenia cases. Our system decomposes evaluation into seven specialized agents with RAG-augmented DSM-5/ICD-11 knowledge: symptom extraction, standard matching, differential diagnosis, temporal analysis, counterfactual reasoning, voting, and arbitration. Evaluated on 71 diagnosed cases, our framework achieves 95.8% PPA (Positive Percent Agreement), outperforming single-agent baseline (70.4%) by 25.4% and single-agent+RAG (87.3%) by 8.5%. Because the dataset is case-only, results reflect case detection rather than screening accuracy; specificity is not estimable. Ablation reveals RAG contributes +16.9% and multi-agent specialization adds +8.5%, demonstrating both are critical for psychiatric symptom-criteria evaluation. Overall, our seven-agent RAG framework structures and traces interview-based symptom-to-criterion assessment and offers a scalable automated prototype; clinical use still requires multi-center validation with controls and first-episode cases to confirm specificity and external generalizability. pubtype: Academic Journal doctype: equations & formulas proceedings research tables/charts Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|