MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202541116024 A) filed by A. Udhaya Veena; J. A. Adlin Layola; V. Gokul & V. Devatarika; Kavitha N; G. Puthillbai; S. Brindha; R. Niruban; and S. Vasudhevan, Chennai, Tamil Nadu, on Nov. 24, 2025, for 'ai-based system for automated detection of anatomical abnormalities in medical images.'
Inventor(s) include A. Udhaya Veena; A. Adlin Layola; V. Gokul & V. Devatarika; Kavitha N; G. Puthillbai; S. Brindha; R. Niruban; and S. Vasudhevan.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "The invention discloses an AI-based system for automated detection of anatomical abnormalities in medical images such as MRI, CT, X-ray, ultrasound, and PET scans. The system comprises a pre-processing module for noise reduction, contrast enhancement, artifact removal, and normalization, followed by a hybrid feature extraction architecture integrating convolutional neural networks for spatial feature learning and transformer encoders for global contextual understanding. An abnormality detection module identifies and segments pathological regions including tumors, lesions, fractures, cysts, hemorrhages, and structural deformities. A classification module generates diagnostic labels and confidence scores, while an output visualization unit Pr9 cuces heatmaps and interpretable overlays for clinical validation. The system optionlly incorporates federated learning to allow secure model improvement withot!t sharing sensitive patient data. The invention provides accurate, real-time, and. explainable computer-aided diagnosis, improving clinical decision-making apd reducing radiologist workload."
Disclaimer: Curated by HT Syndication.
