MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202621048309 A) filed by Symbiosis International Deemed University on April 15, 2026, for A Machine Learning Based System And Method For Predictive And Early Detection Of Mental Health Disorders.
Inventors include Rajat Jain; Akash Ujawane; Divyanshu Dharmik; and Prof. Rajeshwar Balla.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: ABSTRACT A MACHINE LEARNING BASED SYSTEM AND METHOD FOR PREDICTIVE AND EARLY DETECTION OF MENTAL HEALTH DISORDERS The present invention provides a machine learning based system (100) and method for predictive and early detection of mental health disorders using multi-source data integration. The system comprises a data acquisition module (110) for collecting data from text sources (111), speech (112), social media (113), and wearable sensors (114); a data preprocessing module (120) with normalization unit (121) and feature extraction unit (122); a machine learning classification module (130) comprising a BERT model (131) achieving 94% accuracy and XGBoost model (132) achieving 81% accuracy integrated through an ensemble integration unit (133); and an output module (140) providing classification results (141), alerts (142), and personalized recommendations (143). The system classifies mental health conditions including Normal, Depression, Suicidal ideation, Anxiety, Stress, Bi-Polar disorder, and Personality Disorder with high accuracy through transformer-based natural language processing and gradient boosting algorithms, enabling early detection and timely intervention for improved patient outcomes. [
Disclaimer: Curated by HT Syndication.
