Prediction of suicidal ideation through a facial expression reading algorithm in patients with depression

• Grant number: 25/11977-2
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: August 01, 2025
• End date: June 30, 2026
• Field of knowledge: Health Sciences - Collective Health - Preventive Medicine
• Principal Investigator: Paulo Rossi Menezes
• Grantee: Lucas Alvarenga Janz
• Host Institution: Faculdade de Medicina (FM). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 21/12901-9 - National Center for Research and Innovation in Mental Health (CISM), AP.ESP

Abstract

Depressive Disorder is one of the most prevalent psychiatric conditions, characterized by recurrent episodes and mood fluctuations over time, which poses a challenge to the implementation of effective therapeutic strategies. In this context, the National Center for Research and Innovation in Mental Health (CISM) at the University of São Paulo (USP) is conducting a study aimed at predicting suicidal ideation through an algorithm capable of interpreting facial expressions in patients with depression. The objective is to analyze the correlation between the Dysthymic Index (DI), obtained from facial images, and the scores from the Patient Health Questionnaire-9 (PHQ-9).This is an observational and longitudinal study, lasting three months, involving patients diagnosed with moderate to severe depression who are followed at the Psychosocial Care Center (CAPS) in Jaguariúna. Participants will be invited to join the study during their regular consultations and, upon signing the Informed Consent Form, will complete a sociodemographic questionnaire, take daily selfies at random times (morning, afternoon, and evening), and fill out the PHQ-9 on a weekly basis.The participating student will play a key role in the development of the study, maintaining direct contact with the CAPS, conducting regular visits to ensure participant adherence and compliance with research protocols. The student will also assist in data analysis, particularly in examining the relationship between the DI and depressive symptoms through accuracy testing, Receiver Operating Characteristic (ROC) curve analysis (AUC), and Pearson correlation, to assess the DI's potential as a biomarker.Throughout the 90-day study period, both the participants and the CAPS team will receive ongoing support from the research team, including weekly meetings to monitor progress. The student will also be responsible for preparing weekly reports on the study's progress, ensuring that data collection is carried out in a systematic and organized manner.The Ego2Save digital tool is expected to provide a more accurate and accessible means of monitoring depressive symptoms, promoting better treatment adherence and enabling more timely clinical interventions. The project represents a significant advancement in the application of digital technologies in mental health, contributing to innovation in the detection and follow-up of depressive disorders. (AU)