MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202641071397 A) filed by Sona College Of Technology on June 09, 2026, for Machine Learning-Based System For Stroke Risk Prediction Using Patient Health Records.
Inventors include Subashini S M C; Hemalatha K; Vaishnavi Dali A; Anbarasi K; Devika T; and Divya C.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: ABSTRACT The present invention relates to a machine-learning-based stroke risk prediction system and method configured to estimate the possibility of future stroke occurrence using patient demographic, clinical, and lifestyle data. The system comprises a data input module, a preprocessing module, a feature engineering and selection module, a data balancing module employing Synthetic Minority Oversampling Technique (SMOTE), and a machine learning prediction module utilizing ensemble learning algorithms including Random Forest, XGBoost, and voting classifier models. The invention addresses class imbalance in medical datasets and improves identification of under-represented high-risk stroke cases. A model evaluation and output module generates probability-based low-risk or high-risk stroke indications using performance metrics including accuracy, precision, recall, F1-score, and AUC. The invention functions as a preventive healthcare decision-support framework for assisting healthcare professionals in early intervention, risk prioritization, and improved clinical decision-making.
Disclaimer: Curated by HT Syndication.
