MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541124392 A) filed by B V Raju Institute Of Technology, Narsapur, Telangana, on Dec. 10, 2025, for 'adaptive multi-paradigm ensemble system for automated breast cancer classification.'
Inventor(s) include C. Naga Swaroopa; G Geetha; Ashok Kumar Nanda; and V Pradeep Kumar.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "This invention relates to AI-based medical diagnostics, particularly to a system that uses multiple machine learning paradigms and ensemble techniques to automatically classify breast cancer from medical images and clinical datasets. It belongs to the fields of medical image analysis, computational healthcare, and computer-aided diagnosis (CAD) 3. BACKGROUND OF THE INVENTION: Breast cancer diagnosis traditionally depends on expert interpretation of mammograms and histopathology images, which is time-consuming and subject to human error. Existing AI solutions often rely on single models that struggle with variability in medical data and inconsistent performance across datasets. There is a need for a more reliable and adaptive classification approach. This invention addresses these limitations by integrating multiple learning paradigms into a unified adaptive ensemble system that improves accuracy, consistency, and robustness in breast cancer detection. 4. OBJECTIVES OF THE INVENTION: The primary objectives of the present invention are: * To design an automated, adaptive breast cancer classification system using multi-paradigm machine learning. * To combine deep learning, supervised learning, and statistical algorithms into an optimized ensemble. * To improve diagnostic accuracy, reduce false negatives, and ensure robust decision support. * To create a system that adapts automatically to dataset characteristics and model performance. * To support clinical workflows through fast, reliable, and scalable automated cancer detection. ________________________________________ 5. SUMMARY OF THE INVENTION: The invention introduces an Adaptive Multi-Paradigm Ensemble System that integrates diverse computational models to classify breast cancer with high precision. The system consists of preprocessing units, feature extractors, a model pool, and an adaptive ensemble engine. Deep learning models extract high-level image features, while supervised learners and statistical classifiers analyze structured and radiomic data. The adaptive ensemble assigns weights dynamically based on individual model performance, dataset quality, and classification difficulty. A decision optimizer evaluates confidence scores and produces a final diagnostic prediction. The system supports early detection, reduces diagnostic variability, and ensures stable performance across heterogeneous environment________________________________________ 6. DETAILED DESCRIPTION OF THE INVENTION: 1.Data Preprocessing Unit: Handles noise removal, normalization, augmentation, and imbalance correction using SMOTE, histogram equalization, and filtering. 2.Feature Extraction Module: CNNs extract image features; traditional ML models extract numerical and clinical features. 3.Model Pool: Includes CNNs, SVM, Random Forest, Gradient Boosting, Naive Bayes, and KNN models. 4.Adaptive Ensemble Engine: Uses stacking, weighted voting, or meta-learning. Weights are automatically recalculated using performance metrics such as F1-score and AUC. 5.Decision Optimizer: Produces final output by evaluating confidence levels, minimizing misclassifications. 6.Deployment Module: Integrates with clinical interfaces or hospital information systems for real-time diagnosis."
Disclaimer: Curated by HT Syndication.
