Intellectual Property India Publishes Patent Application for 'An Advanced Vertically Stacked Deep Learning-Based System For Automated Detection And Severity Classification Of Diabetic Retinopathy' Filed by Podugu Devi Pradeep; Dr. G. Indira Devi; Dr. D. Madhavi; and Anil Neerukonda Institute Of Technology And Sciences

MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202641009034 A) filed by Podugu Devi Pradeep; Dr. G. Indira Devi; Dr. D. Madhavi; and Anil Neerukonda Institute Of Technology And Sciences, Vizianagaram, Andhra Pradesh, on Jan. 29, for 'an advanced vertically stacked deep learning-based system for automated detection and severity classification of diabetic retinopathy.'

Inventor(s) include Dr. G. Indira Devi; and Dr. D. Madhavi.

The application for the patent was published on Feb. 13, under issue no. 07/2026.

According to the abstract released by the Intellectual Property India: "The present invention relates to a computer-implemented system and method for automated detection and severity classification of diabetic retinopathy (DR) using an advanced deep learning framework. The proposed method processes retinal fundus images acquired from publicly available datasets, including the APTOS-2019 dataset. Initially, the input images are subjected to preprocessing using an adaptive Gaussian filtering technique, followed by image quality enhancement employing a Modified Contrast Limited Interval-value Fuzzy Histogram Equalization (MCLI-FHE) method. Subsequently, retinal regions are segmented using a Patch Merging Swin Transformer (PMST) to isolate clinically relevant structures. Feature extraction is performed on the segmented images using Gray Level Dependence Matrix (GLDM) and Gray Level Co-occurrence Matrix (GLCM) techniques to derive discriminative texture and spatial features. The extracted features are then provided to a Vertically Stacked Residual Convolutional Long Short-Term Enhanced Crayfish Squeeze Excitation Network (VSRCL-ECSEN) for classification of diabetic retinopathy severity. A Crayfish Optimization Algorithm (COA) is employed for automated hyper parameter tuning of the deep learning model to improve classification performance. The proposed invention enables accurate identification of retinal lesions and classification of Diabetic retinopathy into no DR, mild, moderate, severe, and proliferative stages. Experimental evaluation on the APTOS-2019 dataset demonstrates high performance in terms of accuracy, precision, sensitivity, specificity, and F1-score. The disclosed system provides an efficient and reliable computer-aided diagnostic tool for early detection and severity assessment of diabetic retinopathy, thereby supporting clinical decision-making and reducing the risk of vision impairment."

Disclaimer: Curated by HT Syndication.