Sensing Psychosis: Using Artificial Intelligence & Machine Learning to Advance Psychiatric Assessment-View Presentation

Video Transcription

Video Summary

In his presentation, Justin Baker, an associate professor of psychiatry at Harvard Medical School, explores the integration of AI and machine learning into psychiatric assessment, emphasizing its potential in improving how mental illnesses are evaluated and treated. Traditionally, assessments rely on distinguishing between control and case groups, which can be complicated due to the spectrum of mental conditions, comorbidities, and developmental stages. Baker discusses the use of latent construct models to devise estimates of unmeasured mental states by employing closed-loop systems. These systems, facilitated by AI, can enhance clinicians' abilities to interpret multiple sensory inputs and inform diagnoses. Baker illustrates the application of AI in clinical settings via three use cases: monitoring inpatient sleep and activity using wearable devices, assessing dyadic interactions via video analysis to understand expressed emotions and cognitive states, and tracking comprehensive patient data longitudinally to correlate symptoms with contextual and objective data. Despite its potential, using AI in psychiatry raises ethical and privacy concerns, which requires careful handling to ensure patient trust and data integrity. The overarching aim is to leverage AI to complement, not replace, psychiatric practice, enhancing precision and individualized care.

Keywords

AI in psychiatry

machine learning

psychiatric assessment

latent construct models

closed-loop systems

wearable devices

ethical concerns

patient trust

individualized care