MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641094731 A) filed by Dhanalakshmi Srinivasan College Of Engineering And Technology on August 05, 2026, for A Hybrid Enhanced Cnn Model For Detection Of Lung Infections Using X-Ray And Ct Modalities.

Inventors include P. R. Ajitha; Balamurugan T; and Gurudeva P.

The application for the patent was published on September 11, 2026, under issue no. 37/2026.

Abstract: A Hybrid Enhanced CNN Model for Detection of Lung Infections Using X-Ray and CT Modalities. The present invention relates to a Hybrid Enhanced Convolutional Neural Network (CNN) Model for the automated detection and classification of lung infections using chest X-ray and Computed Tomography (CT) scan images. The invention integrates advanced image preprocessing techniques, deep learning architectures such as LeNet and ResNet, and Explainable Artificial Intelligence (XAI) methods to provide accurate, reliable, and interpretable diagnostic support. Initially, medical images undergo preprocessing operations including resizing, normalization, noise reduction, and contrast enhancement to improve image quality and model performance.The processed images are then analyzed using hybrid CNN models trained to identify infection-related abnormalities within lung tissues. To enhance transparency and clinical trust, the system employs Gradient-weighted Class Activation Mapping (Grad-CAM), which generates visual heatmaps highlighting the specific lung regions that contribute to the model's predictions. This enables healthcare professionals to better understand and validate the diagnostic outcomes.The invention is further implemented through a Django-based web platform that allows clinicians and medical staff to upload X-ray or CT scan images and receive real-time diagnostic results. By reducing dependence on manual image interpretation, minimizing diagnostic errors, and accelerating the detection process, the proposed system improves clinical efficiency and supports early disease diagnosis. The invention contributes significantly to AI-assisted healthcare by enhancing diagnostic accuracy, accessibility, and patient care outcomes

Disclaimer: Curated by HT Syndication.