MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202641112547 A) filed by Vignans Nirula Institute Of Technology And Science For Women on September 19, 2026, for Automatic Method For Optic Disc Segmentation Using Deep Learning On Retinal Fundus Images.
Inventors include S. Yallamandaiah; Avula Swathi; Syed Nahiya; Mula Swathi; Medida Sumathi; Kambagiri Lavanyasree; Unnava Kasturi; Thota Lavanya; and Narapusetty Khamjakshi.
The application for the patent was published on September 25, 2026, under issue no. 39/2026.
Abstract: Automatic optic disc segmentation using deep learning is a medical image processing technique designed to accurately detect and isolate the optic disc region from retinal fundus images. The optic disc plays a crucial role in diagnosing eye diseases such as glaucoma, as it helps in evaluating structural changes in the retina. The proposed method utilizes convolutional neural networks (CNNs), including models such as MobileNetV2 for region-of-interest detection and U-Net for precise segmentation. By applying preprocessing techniques like image resizing, augmentation, and normalization, the system improves accuracy and robustness. The approach achieves high performance and produces results comparable to ophthalmologist annotations, making it suitable for automated glaucoma screening and clinical decision- support systems.
Disclaimer: Curated by HT Syndication.
