MUMBAI, India, April 17 -- Intellectual Property India has published a patent application (202531129768 A) filed by Narula Institute Of Technology, Kolkata, West Bengal, on Dec. 21, 2025, for 'voice-driven ensemble system for parkinson's disease monitoring.'
Inventor(s) include Anik Baul; Samadrita Karmakar; Srinjay Dasgupta; Prabuddha Chatterjee; Debopriya Dey; Debrupa Pal; Mrinmoy Guria; and Dr. Abhishek Hazra.
The application for the patent was published on April 17, under issue no. 16/2026.
According to the abstract released by the Intellectual Property India: "A computerized system and method for non-invasively predicting the severity of Parkinson's Disease is disclosed, employing a highly accurate multi-stage ensemble machine learning architecture. The system processes acoustical features extracted from a patient's voice, utilizing these vocal biomarkers to estimate motorUPDRS and totalUPDRS scores. The methodology involves a first stage of parallel regression using a Deep Neural Network and a first Random Forest Regressor, the outputs of which are aggregated to yield an intermediate prediction. The novelty resides in a second refinement stage, wherein a meta-level Random Forest Regressor is exclusively trained on the residual errors of the intermediate prediction. This residual learning mechanism precisely adjusts the predictive output, significantly enhancing the accuracy and robustness of the severity quantification. The invention thus provides an objective, rapid, and cost-effective diagnostic tool suitable for remote telemonitoring, enabling timely clinical intervention and superior patient management."
Disclaimer: Curated by HT Syndication.
