Study Using Multi-Modal Sentiment Analysis to Predict Recurrence Risk in Bipolar Disorder Observational Study Including Patients and Healthy Controls Aged 18 to 64

What this Trial Is About

Researchers are investigating bipolar disorder BD, a complex mental health condition with challenging treatment and poor prognosis. This study aims to improve the prediction of BD recurrence by using advanced sentiment analysis technology that combines multiple types of data, such as text, audio, visual cues, eye movements, and electrophysiology. The goal is to develop biological markers and an algorithm model for better risk prediction and to support future clinical decision-making in BD management. The study will enroll 200 participants diagnosed with bipolar disorder according to DSM-5 criteria and 200 healthy controls. Participants will be observed without any intervention, with the collection of multi-modal data to analyze emotional expressions and identify recurrence risk. The research builds on previous neuroimaging and big data studies and uses a prospective cohort design to verify and optimize the prediction model over time. Participants with bipolar disorder will be followed for one year, with assessments every three months to monitor recurrence status, while healthy controls will be assessed only once at baseline. The study involves various evaluations including visual, auditory, and emotional data collection to support the algorithm development. This observational study seeks to offer new methods for predicting bipolar disorder recurrence and enhancing clinical diagnosis and treatment approaches.

An Exploratory Study on the Prediction of Recurrence Risk of Bipolar Disorder Using Sentiment Analysis Technology Based on Multi-modal Feature Fusion