MUMBAI, India, Feb. 27 -- Intellectual Property India has published a patent application (202641017623 A) filed by Swarna Bharathi Institute Of Science And Technology, Khammam, Telangana, on Feb. 17, for 'ensemble deep learning framework for multi-class diabetes risk stratification systems.'
Inventor(s) include Mr. Ananthala Mallikarjun; Mrs. Cherukuri Krishnaveni; Tigulla Vidyasri; Janapati Sri Ankitha; and Chirakala Praneetha.
The application for the patent was published on Feb. 27, under issue no. 09/2026.
According to the abstract released by the Intellectual Property India: "The present invention discloses an ensemble deep learning framework for multi-class diabetes risk stratification. The system integrates heterogeneous deep neural networks, including convolutional, recurrent, and fully connected architectures, to analyze clinical, biochemical, demographic, and lifestyle data. A preprocessing module performs data normalization, missing value imputation, and class balancing to address imbalanced risk categories. Feature-level fusion and adaptive weighted ensemble mechanisms are employed to enhance classification robustness and generalization. The framework stratifies individuals into multiple risk levels such as non-diabetic, pre-diabetic, controlled diabetic, and high-risk diabetic, enabling early intervention and personalized management. A dynamic optimization layer tunes model parameters using meta-learning strategies to improve predictive accuracy. The proposed system supports cloud-based deployment with privacy-preserving mechanisms to protect sensitive health data. The invention significantly improves multi-class risk prediction accuracy, interpretability, and scalability for clinical decision support systems in diabetes management."
Disclaimer: Curated by HT Syndication.
