Dimensional symptom scales are sensitive measures of the course of psychosis. They are the gold-standard tools for routinely monitoring psychotic symptoms, but their use is inconsistent at best: fewer than 30% of psychiatrists report even occasionally using them, while studies often differ substantially in which and how many scales they assess, even when measuring the same symptoms. Thus, most people with psychotic disorders, whether in the clinic or in research, have their experiences only partially translated into quantitative dimensions, if at all.
A common objection to dimensional scales is that they are lengthy to complete, and our project seeks to develop a pragmatic alternative. We will create a system (SYLFAEN-PSY), informed by lived experience and based on large language models, to generate proxy measurements of psychotic symptoms by integrating multimodal data. These proxies will be validated against symptom scales in deeply characterised cohorts of individuals with psychotic disorders, and in research data from mental health services. Through extensive assessments of uncertainty and bias, we aim to develop a new standard for AI-based information extraction and inference in mental health research, providing a way to assess the experiences of people with psychosis at scale, and removing a long-standing barrier to symptom-based research."
Foundation models · multimodal data · LLM-based information extraction and inference
Psychosis · symptom measurement