Intellectual Property India Publishes Patent Application for A Deep Learning-Based Medical Imaging Framework For Automated Disease Diagnosis And Clinical Decision Support

MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202641125043 A) filed by Dr. L Anand; and Dr. L. N. B Srinivas on October 19, 2026, for A Deep Learning-Based Medical Imaging Framework For Automated Disease Diagnosis And Clinical Decision Support.

Inventors include Dr. L Anand; and Dr. L. N. B Srinivas.

The application for the patent was published on October 30, 2026, under issue no. 44/2026.

Abstract: A deep learning-based medical imaging framework and method for automated disease diagnosis and clinical decision support is disclosed. An image-ingestion and pre- processing module receives a medical image of at least one modality, comprising radiographic, computed-tomography, magnetic-resonance, ultrasound, digital- pathology, or ophthalmic images, and produces a standardized representation, the pre-processing reducing source-dependent differences. A deep-learning analysis module comprising at least one trained deep-learning model analyses the representation so as to detect and characterize an indication of a disease and to produce a disease-indication output, the module being trained for robustness across sources and populations. A localization and explanation module localizes the responsible region, by segmentation or a class-discriminative activation map, and explains the output; a confidence-estimation module produces a calibrated confidence and an uncertainty; and an out-of-distribution facility withholds a confident output and flags unfamiliar images for review. A clinical decision-support module generates decision-support information, prioritizes urgent cases, and presents the information to a clinician through a clinical information system. The framework furnishes robust, localized, confidence-qualified, and explainable decision support to assist the clinician, without diagnosing in substitution for the clinician or administering treatment, so as to improve the speed, consistency, and availability of the interpretation of medical images.

Disclaimer: Curated by HT Syndication.