MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202621048315 A) filed by Symbiosis International Deemed University on April 15, 2026, for Machine Learning Based Diabetes Prediction System Using Ensemble Classification Algorithms And Multi-Parameter Health Data Analysis.
Inventors include Rishitha Bajaj; Shreya Gholase; Nandini Ashtankar; and Dr. Priya Dasarwar.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: ABSTRACT MACHINE LEARNING BASED DIABETES PREDICTION SYSTEM USING ENSEMBLE CLASSIFICATION ALGORITHMS AND MULTI- PARAMETER HEALTH DATA ANALYSIS The present invention discloses a machine learning based diabetes prediction system (100) for early detection and risk assessment of diabetes mellitus. The system comprises a data acquisition module (110) for receiving multi-parameter health data including glucose, blood pressure, BMI, insulin, skin thickness, diabetes pedigree function, pregnancies, and age parameters. A preprocessing module (120) handles missing values and removes outliers using Z-score and IQR methods. A feature engineering module (130) performs feature scaling, class imbalance handling using SMOTE, and data splitting. A classification module (140) implements five machine learning algorithms including Logistic Regression, Decision Tree, Random Forest, Support Vector Classifier, and K-Nearest Neighbors. A model evaluation module (150) assesses performance using accuracy, precision, recall, F1-score, and confusion matrix analysis. The Support Vector Classifier achieves highest accuracy of 82 percent, enabling reliable early diabetes detection. [
Disclaimer: Curated by HT Syndication.
