MUMBAI, India, April 17 -- Intellectual Property India has published a patent application (202531132752 A) filed by Asansol Engineering College; Jis University; and Abacus Institute Of Engineering & Management, Asansol, West Bengal, on Dec. 28, 2025, for 'attention-augmented vgg16 for improved retinal image processing and analysis.'
Inventor(s) include Sumanta Karmakar; Dr. Jyotirmoy Chatterjee; Dr. Sambit S Mondal; Mrs. Lumbini Bhowmik; Dr. Kailash Pati Mondal; and Dr. Goutam Kumar Das.
The application for the patent was published on April 17, under issue no. 16/2026.
According to the abstract released by the Intellectual Property India: "Precise analysis of retinal images is essential for the early identification and diagnosis of diseases that threaten vision. In practice, the quality of fundus images is often compromised by issues such as low contrast, noise, and uneven illumination, which hinder both manual evaluation and automated diagnostic systems. To address these challenges, this study introduces a deep learning-driven enhancement framework that integrates the VGG16 convolutional architecture with Squeeze-and-Excitation (SE) attention modules. The proposed SE-VGG16 network adaptively emphasizes informative channel-wise features, thereby enhancing the visibility of critical retinal structures-such as blood vessels and optic disc contours-while attenuating irrelevant background details. Experiments conducted on benchmark datasets (DRIVE, STARE, and CHASE) demonstrate the effectiveness of the approach, achieving segmentation accuracies exceeding 96% and AUC values above 93%. Further evaluation using PSNR, SSIM, and entropy metrics confirms significant improvements in both image clarity and structural fidelity. Compared to conventional and state-of-the-art methods, the proposed framework consistently yields higher F1-scores, sensitivity, and specificity, highlighting its strength as a robust retinal image enhancement technique and a valuable pre-processing step for downstream automated ophthalmic analysis."
Disclaimer: Curated by HT Syndication.
