MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202641008991 A) filed by Vemu Institute Of Technology, Chittoor, Andhra Pradesh, on Jan. 29, for 'a predictive diagnostic framework for early detection of diabetes using integrated machine learning classification models.'

Inventor(s) include Mr. Karamala Naveen; Mr. Velakaturi Lakshmi Prasad; and Dr T Haritha.

The application for the patent was published on Feb. 13, under issue no. 07/2026.

According to the abstract released by the Intellectual Property India: "This invention presents a machine learning-driven system for prognosis and diagnosis of diabetes to enhance early detection and predictive precision. Diabetes mellitus is a long-term disease with increasing global incidence, and early diagnosis is paramount to avoiding serious complications. The system relies on structured patient information and implements a multi-model pipeline using classifiers including SVM, Logistic Regression, Random Forest, AdaBoost, Naive Bayes, XGBoost, CatBoost, and MLP. The approach starts with comprehensive preprocessing, involving outlier rejection, missing value imputation, feature scaling, and feature selection. The models are trained with stratified K-fold cross-validation and optimized through grid search. A strategy of ensemble using soft weighted voting improves prediction credibility, with AUC values used to determine model weights. Patients interact through a web-based interface to upload data, train models, enter symptoms, and get diagnostic results. The system also provides prognosis by predicting future diabetic risk based on past trends. It is architected for deployment into clinics, research labs, and telemedicine platforms. This innovation provides a scalable, accurate, and interpretable diabetes management solution that combines the use of advanced analytics with everyday practicality. It empowers healthcare providers and patients alike with timely insights, leading to better outcomes and more effective healthcare delivery."

Disclaimer: Curated by HT Syndication.