Intellectual Property India Publishes Patent Application for A System And Method For Automated Medical Image Segmentation Using Machine Learning-Based Anatomical Boundary Refinement

MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641069301 A) filed by Dr. K. B. Venkata Brahma Rao; Ms. Shruthi K. N.; Dr. Lokabhiram Dwarakanath; Dayananda Sagar Academy Of Technology And Management; Dr. M. Narendra; Dr. Vijaya Kumar A. V.; Dr. Prabhakar M.; Dr. Sushant Savita Madhukar Gandhi; Dr. Anupama Shrivastava; and Dr. Amitava Biswas on June 02, 2026, for A System And Method For Automated Medical Image Segmentation Using Machine Learning-Based Anatomical Boundary Refinement.

Inventors include Dr. K. B. Venkata Brahma Rao; Ms. Shruthi K. N.; Dr. Lokabhiram Dwarakanath; Dayananda Sagar Academy Of Technology And Management; Dr. M. Narendra; Dr. Vijaya Kumar A. V.; Dr. Prabhakar M.; Dr. Sushant Savita Madhukar Gandhi; Dr. Anupama Shrivastava; and Dr. Amitava Biswas.

The application for the patent was published on June 12, 2026, under issue no. 24/2026.

Abstract: The present invention relates to a system and method for automated medical image segmentation using machine learning-based anatomical boundary refinement. The system receives medical image data from one or more imaging modalities and preprocesses the data through normalization, enhancement, artifact suppression, and spatial alignment. A machine learning segmentation engine generates an initial segmentation mask for anatomical or pathological regions of interest. An anatomical boundary refinement module corrects segmentation errors by applying boundary continuity, topology correction, anatomical adjacency, shape regularization, and spatial consistency rules. A confidence estimation module generates confidence information for reliable and uncertain mask regions, while a clinical validation module verifies anatomical and clinical consistency of the refined segmentation output. The system generates a final segmentation output including a segmentation mask, boundary overlay, confidence map, quantitative measurements, metadata, and editable clinical review data. The invention improves segmentation accuracy, reduces manual workload, enhances clinical reliability, and supports standardized medical imaging workflows. Accompanied Drawing [FIGS. 1-2]

Disclaimer: Curated by HT Syndication.