MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202611073230 A) filed by Poornima University; Mr. Rocky Kumar; and Mr. Gaurav Kumar on June 12, 2026, for System And Method For Clinical Decision Support Through Probabilistic Federated Learning Of Iot- Derived Physiological Signals.
Inventors include Mr. Rocky Kumar; and Mr. Gaurav Kumar.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: The present invention relates to a hardware-integrated clinical intelligence system configured for real-time, uncertainty-aware medical decision making using physiological data acquired from multiple sensing units. The system comprises an edge processing unit with a probabilistic inference processor that generates clinical predictions in the form of distribution-based outputs along with associated uncertainty measures. A signal reliability computation mechanism evaluates the quality of acquired physiological signals and integrates such reliability parameters into the decision-making process. An adaptive decision processing unit combines predicted clinical risk, uncertainty values, and signal reliability to generate a composite decision score, which is used to produce graded clinical alerts. The system further includes a secure communication unit configured to transmit encrypted model parameters for federated aggregation without sharing raw patient data. By enabling localized processing, uncertainty-driven reasoning, and privacy-preserving collaborative learning, the invention significantly reduces latency, enhances decision reliability, and minimizes false clinical alerts in healthcare environments.
Disclaimer: Curated by HT Syndication.
