MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641085534 A) filed by Manipal Academy Of Higher Education on July 13, 2026, for Secure And Interpretable Machine Learning System And Method For Thermography-Based Body Temperature Prediction.
Inventors include Omkar Prabhu; Snehal Samanth; and Krishnaraj Chadaga.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: Embodiments of the present disclosure relate to an intelligent non-contact body temperature prediction system (100) and method (200). The system (100) includes a data acquisition unit (102) is configured to receive infrared thermographic measurements. The optimization unit (108) is configured to automatically optimize model parameters using one or more search techniques. An explainable artificial intelligence (XAI) unit (110) generates global and local interpretability outputs. A security unit (112) protects physiological data, prediction results, and model parameters through authenticated encryption and integrity verification. An output unit (114) generates a predicted body temperature together with confidence metrics, interpretability indicators, and secured prediction data for presentation through a clinical decision support interface. The system (100) utilizes facial thermal measurements, demographic information, environmental measurements, and derived thermal features to provide reliable non-contact fever screening, continuous patient monitoring, and secure healthcare analytics while improving prediction accuracy, model transparency, and protection of sensitive medical information.
Disclaimer: Curated by HT Syndication.
