Integrating conversational AI for accurate psychiatric diagnosis in primary care

This research project will explore how advanced artificial intelligence technology, specifically large language models (LLMs), can improve the Mental Health Assessment Tool (MHAT) developed by Harrison Healthcare. MHAT is a secure online tool designed to help primary care doctors identify and diagnose mental health conditions based on structured questionnaires and clinical guidelines. The project will test whether adding LLMs to MHAT can make the tool more effective by providing more personalized, empathetic, and flexible assessments. This improvement could lead to better patient experiences and more accurate mental health diagnoses.

A doctoral intern will work with Harrison Healthcare over a 12-month period to integrate and test the enhanced MHAT. The expected benefits include reducing the time doctors spend on assessments, improving the accuracy of mental health diagnoses, and helping patients feel more heard and supported during their care. This work will also position Harrison Healthcare as a leader in digital mental health innovation while contributing to the broader field of trustworthy AI in healthcare.

Faculty Supervisor:

Ga Wu

Student:

Partner:

Harrison Healthcare

Discipline:

Computer science

Sector:

Health and Related Sciences & Technology

University:

Dalhousie University

Program: