MUMBAI, India, Oct. 26 -- Intellectual Property India has published a patent application (202611111904 A) filed by Krishna Institute Of Engineering & Technology Kiet on September 17, 2026, for Unified Multi-Disease Risk Prediction And Clinical Reasoning System With Retrieval-Grounded Symptom Analysis.
Inventors include Arushi Singh; Ritika Singh; Shubham Yadav; Shantanu Yadav; and Shruti Yadav.
The application for the patent was published on October 23, 2026, under issue no. 43/2026.
Abstract: Disclosed is a processor-executed clinical decision support system and method for multi-disease risk screening and reasoning through a single interface. Structured biomarker inputs are applied to disease-specific machine learning classifiers and a generalized symptom classifier to produce a probabilistic prediction. The prediction output, comprising a predicted condition, a probability and a summary of the inputs, is transformed into a structured prompt that conditions a large language model to return structured clinical reasoning, including agreement or disagreement with the prediction, a differential diagnosis, a recommended diagnostic test, a specialist referral and a lifestyle suggestion. For free-text symptom input, a retriever retrieves relevant chunks from a curated medical knowledge base and a severity-scoring module computes a symptom-severity measure, and the large language model generates an interpretation grounded in the retrieved chunks and the severity measure. The system further generates condition-specific dietary guidance.
Disclaimer: Curated by HT Syndication.
