NCT07875816|Active Not Recruiting
LLM Psychiatric Intake AI Clinical Performance Study
Development and Evaluation of an LLM-Based, Clinician-in-the-Loop Intake System for Mental Health
Loading search results
Development and Evaluation of an LLM-Based, Clinician-in-the-Loop Intake System for Mental Health
Sponsor study identifier
interventional
120
1 country
1
Study Start
November 7, 2025
Actual (reported)Primary Completion
February 27, 2027
EstimatedPlanned date reported by the registry; this milestone is not confirmed as achieved.
Study Completion
November 6, 2027
EstimatedPlanned date reported by the registry; this milestone is not confirmed as achieved.
This study will evaluate the clinical performance and applicability of a large language model-based psychiatric intake support system for caregivers of patients undergoing an initial or re-intake psychiatric evaluation. Caregivers will complete a single adaptive AI interview remotely before the scheduled outpatient visit. The system will generate a structured clinician-facing report and diagnostic screening candidates for nine prespecified target conditions. Treating clinicians will review the AI-generated information while retaining full authority over the final clinical assessment and diagnosis. The system will be evaluated in terms of diagnostic-assistance performance, interview and information-collection performance, documentation quality, acceptability and user experience, and safety.
Registry-reported information. Elapsed time and enrollment totals alone do not measure study progress or recruitment pace.
November 6, 2027
EstimatedPlanned date reported by the registry; this milestone is not confirmed as achieved.
Click on a node to explore related trials.
Dates and qualifiers reported by the registry. Estimated dates do not confirm that a milestone has occurred.
November 7, 2025
Actual (reported)February 10, 2026
Actual (reported)October 15, 2026
Actual (reported)February 27, 2027
EstimatedPlanned date reported by the registry; this milestone is not confirmed as achieved.
November 6, 2027
EstimatedPlanned date reported by the registry; this milestone is not confirmed as achieved.
Pooled sensitivity of AI-generated diagnostic candidates for the nine target disorders
The pooled sensitivity of the AI-generated diagnostic candidates will be evaluated against the reference primary diagnosis determined by a psychiatrist based on clinical information accumulated during the 8 weeks following the initial clinical visit. Among participants whose reference primary diagnosis corresponds to one of the nine target disorders, a true-positive case is defined as one in which the AI-generated candidate list includes the reference diagnosis. A false-negative case is defined as one in which the AI does not include the reference diagnosis or returns a deferred decision. Pooled sensitivity is calculated as the number of true-positive cases divided by the total number of participants whose reference primary diagnosis is one of the nine target disorders. The outcome ranges from 0 to 1, with higher values indicating better diagnostic candidate-generation performance.
Up to 8 weeks after the day 1
Disorder-specific sensitivity of AI-generated diagnostic candidates for each of the nine target disorders
Caregivers complete an approximately 30-minute adaptive LLM-based AI psychiatric intake interview before/around the outpatient visit. The AI system generates a structured intake report for the treating clinician. Experts rate de-identified AI interview logs/reports. Treating clinicians review the AI report in routine care and provide qualitative feedback.
Caregivers complete a single adaptive LLM-based psychiatric intake interview remotely before the patient's scheduled initial or re-intake visit. The web-based interview is accessible by smartphone or tablet and lasts up to 30 minutes. Guided by K-SADS-PL and DSM-5-based clinical domains, the system adaptively explores information typically collected from caregivers during an initial psychiatric assessment. The system generates a structured clinician-facing report containing an interview summary, diagnostic screening candidates for nine prespecified target conditions, safety-related findings, and areas requiring further clarification. The system incorporates mental health-specific safety guardrails tailored to the developmental characteristics and safety needs of children and adolescents. These guardrails restrict diagnostic and treatment advice and detect potential risks, including self-harm, harm to others, suicidal history, substance misuse, psychotic symptoms, and suspected child ab
Gangnam Severance Hospital
Seoul, Gangnam-gu, 06273, South Korea
Jung Y, Vo THA, Moon HS, Choi JY, Oh H, Lee U, Kim E, Lee TY, Lee U. Toward Flexible Psychiatric History-Taking and Visualization: Exploring Clinician Perspectives with Large Language Models. Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems. 2026; Article 503:1-23. doi:10.1145/3772318.3790970.
First Submitted
February 10, 2026
First Posted
October 15, 2026
Study Start
November 7, 2025
Actual (reported)AgingUP: A Pilot Program to Reduce Self-Ageism and Self-Stigma in Older Adults With Severe Mental Illness
Up to 8 weeks after the day 1
Specificity of AI-generated diagnostic candidates for identifying cases outside the nine target disorders
Up to 8 weeks after the day 1
Clinical information-gathering and interviewing performance assessed by the modified Mini-CEX
Within 1 month after completion of caregiver data collection
Information-elicitation and communication performance assessed by the SEGUE Framework
Within 1 month after completion of caregiver data collection
Clinical summarization and documentation quality assessed by the PDSQI-9
Within 1 month after completion of caregiver data collection
Counseling and relational competence
Up to 2-month after Day 1
Behavioral self-disclosure
Day 1
Relational empathy
Day 1
Contact the study team to confirm eligibility.
Primary Completion
February 27, 2027
EstimatedPlanned date reported by the registry; this milestone is not confirmed as achieved.
Study Completion
November 6, 2027
EstimatedPlanned date reported by the registry; this milestone is not confirmed as achieved.
Last Updated
October 15, 2026
Record last verified: 2026-09
Download the complete ClinicalTrials.gov record imported by Syfrah, including fields not shown in this page. The registry may contain a newer revision.
Source: ClinicalTrials.gov.
JSON downloads include source attribution and retrieval details alongside the unchanged original record or the separately normalized values.
Source retrieved: