MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641080028 A) filed by Srm University-Ap on June 29, 2026, for System And Method For Lung Ailment Diagnosis Through Images Using Hybrid Machine Learning Architecture.
Inventors include Dr. Elakkiya Ezhilarasan; Sandeep Reddy Panyala; Dr. Antony Raj Sowridoss; and Dr. S. Priya.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: The present invention relates to a system and method for lung ailment diagnosis through chest radiograph and computed tomography (CT) images using hybrid machine learning architecture.The system comprises of a computing unit (100) comprising one or more processors coupled with dedicated memory and the processor coupled with memories of the computing unit (100) are configured for implementing functionalities of storing chest radiograph and computed tomography (CT) images, implementing Contrastive Language-Image Pre-training (CLIP)-based validator, implementing Hybrid EfficientNetV2–Swin Transformer cross-attention architecture, implementing Quality-Weighted Aggregation (QWA) with differential privacy architecture, implementing Elastic Weight Consolidation (EWC) based continual federated learning (EWC-FL) architecture, implementing Gradient Weighted Class Activation Mapping (GradCAM) driven structured XAI report engine and implementing Entropy-Based Escalation – Deployment model into a single deployable clinical platform.
Disclaimer: Curated by HT Syndication.
