MUMBAI, India, July 13 -- Intellectual Property India has published a patent application (202641081655 A) filed by Ms. N. V. N Sowjanya; Dr. Srikanth Kadainiti; Mr. Aluru Ramesh Khanna; Dr. Aparna Srikantam; Dr. Maithili Devireddy; Ms. P Swetha; Mrs. V. Sravanthi; Mrs. Spurthi Kalevar; Dr. Kishore Kumar Kalevar; and Mr. Hari Kishore Chejarla. on July 02, 2026, for An Artificial Intelligence-Based Method For Early Disease Diagnosis Using Medical Imaging.
Inventors include Ms. N. V. N Sowjanya; Dr. Srikanth Kadainiti; Mr. Aluru Ramesh Khanna; Dr. Aparna Srikantam; Dr. Maithili Devireddy; Ms. P Swetha; Mrs. V. Sravanthi; Mrs. Spurthi Kalevar; Dr. Kishore Kumar Kalevar; and Mr. Hari Kishore Chejarla..
The application for the patent was published on July 10, 2026, under issue no. 28/2026.
Abstract: ABSTRACT The present disclosure relates to an artificial intelligence (AI)-based method and system for early disease diagnosis using medical imaging. The invention utilizes advanced image preprocessing techniques, deep learning architectures, automated feature extraction, and intelligent disease classification to identify pathological abnormalities from medical images at an early stage. The proposed system accepts medical images acquired from multiple imaging modalities including X-ray, Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Ultrasound, and fundus imaging. Initially, acquired medical images undergo preprocessing operations comprising noise reduction, normalization, contrast enhancement, image resizing, artifact removal, and intensity correction. The processed images are subsequently subjected to automated organ and lesion segmentation using a convolutional neural network or transformer-based segmentation model. Extracted image regions are provided to a hybrid artificial intelligence model incorporating convolutional neural networks, attention mechanisms, and transformer-based feature learning for automatic extraction of discriminative disease-related features. The extracted features are classified into predefined disease categories using an optimized deep neural network that generates probability scores representing disease confidence levels. The system further integrates explainable artificial intelligence techniques to produce visual attention maps highlighting abnormal anatomical regions, thereby improving interpretability for clinicians. Diagnostic reports including disease probability, severity estimation, confidence score, and recommended clinical actions are automatically generated. The proposed invention provides improved diagnostic accuracy, reduced false-positive and false-negative rates, shorter diagnosis time, scalability across multiple disease types, and compatibility with hospital information systems and cloud-based healthcare platforms.
Disclaimer: Curated by HT Syndication.
