Dialogue System for Early Mental Illness Detection: Toward a Digital Twin Solution

Abstract

Mental health disorder rates have increased in recent years. In this research, we aim to address the barriers of stigma, accessibility, and affordability in mental healthcare by designing and developing a dialogue system that analyses the mental status of individuals. Additionally, it gives them personalized feedback based on the severity of the mental health problem. We propose a framework based on the concept of a Digital Twin for mental health, which incorporates recent advancements in technology to assess and classify the severity of mental health problems. The chatbot framework was designed in collaboration with a clinical psychiatrist and utilizes pre-trained BERT models, fine-tuned on the E-DAIC dataset, for the detection of various severity levels. The results of this study demonstrate the potential for our method to accurately detect signs of mental health problems with 69% accuracy, and high acceptability and usability with a score of 84.75%.