MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641074918 A) filed by Sasi Institute Of Technology & Engineering; A N V K Swarupa; A Durga Bhavani; K Sandeep; and I Usha on June 17, 2026, for A Cloud Computing–driven Federated Yolov8 Framework For Real-Time Medicinal Plant Disease Detection.
Inventors include Sasi Institute Of Technology & Engineering; A N V K Swarupa; A Durga Bhavani; K Sandeep; and I Usha.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: The proposed invention pertains to an artificial intelligence framework for oral cancer detection with multimodal images and clinical data. The system comprises an oral image acquisition module and clinical symptom acquisition module for acquiring patient information. The acquired images are then passed to a deep feature extraction module (EfficientNet-B0) and the clinical data is preprocessed and fused with the image features by a feature fusion module. The fused features are further optimized by reducing the dimensionality of the features using Principal Component Analysis (PCA). Optimized feature set is used to classify the oral cancer conditions using an XGBoost classification module. An explainable artificial intelligence module (Grad-CAM and SHAP) is used to give visual and feature-level explanations of the predictions to enhance the interpretability of the predictions. Last but not least, a recommendation generation module recommends suitable medical advice for the classification output. The proposed framework enhances the diagnostic accuracy, aids in early detection of oral cancer and offers clarity and understanding in decision making for clinical support.
Disclaimer: Curated by HT Syndication.
