# Prototype of a multimodal AI system for vitiligo detection and mental health monitoring

## TL;DR

This paper introduces an AI system combining lesion detection and mental health monitoring for vitiligo, aiming to improve early diagnosis and patient care.

## Contribution

A novel multimodal AI framework integrating YOLOv11 and BERT for vitiligo detection and mental health assessment.

## Key findings

- YOLOv11 achieved 98.8% mAP for lesion detection.
- BERT-based classifier reached F1 = 0.83 for mental health monitoring.
- Fusion score model achieved AUC = 0.82 for identifying high-risk patients.

## Abstract

Vitiligo is a chronic autoimmune disorder with profound psychosocial implications.

The paper propose a multimodal artificial intelligence (AI) framework that combines and integrates YOLOv11 for the detection of dermatological lesion and a BERT-based sentiment classifier for the monitoring of mental health, supported by questionnaire data sets (DLQI, RSE).

YOLOv11 achieved mAP = 98.8%, precision = 95.6%, recall = 97.0%; the mental health module uses a BERT-based sentiment classifier, fine-tuned in the GoEmotions corpus, reaching F1 = 0.83. A simulated fusion score that integrates the Dermatology Life Quality Index (DLQI) and Rosenberg Self-Esteem (RSE) scores, resulting in an area under the ROC curve (AUC) of 0.82 for the identification of high-risk patients.

The implemented prototype establishes the feasibility of AI-assisted psychodermatology, allowing early diagnosis, emotional monitoring, and real-time alerting by physicians.