Intellectual Property India Publishes Patent Application for Multi-Chest Disease Detection Using Neural Network

MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202641113481 A) filed by Anurag University on September 22, 2026, for Multi-Chest Disease Detection Using Neural Network.

Inventors include Dr. G. Balakrishna; Guna Jashwanthi; Kuchipudi Sairam; and Charugundla Rithwik.

The application for the patent was published on September 25, 2026, under issue no. 39/2026.

Abstract: ABSTRACT The present invention relates to a system (100) for classification of chest X-ray images, comprising an image input module (102), a preprocessing module (102a, 102b), a segmentation module (104), a feature extraction module (106), and a classification module (108). The image input module (102) receives a chest X-ray image, while the preprocessing modules (102a, 102b) enhance the image using an adaptive guided multi-layer side-window box filtering technique. The segmentation module (104) performs lung-region segmentation using a lightweight multi-SegNet architecture. The feature extraction module (106) extracts features from the segmented lung region using an S-transform and Fast Discrete Orthonormal Transform (FDOT). The classification module (108) comprises a Multi-scale Alignment Graph Capsule Binary Light Spectrum Dual Attention Neural Network (MACBDN) incorporating multi-scale alignment graph capsules, binary light spectrum optimization, and dual attention mechanisms to classify the chest X-ray image into normal, COVID-19, viral pneumonia, or bacterial pneumonia categories. The figure associated with abstract is Fig. 1.

Disclaimer: Curated by HT Syndication.