MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202621047073 A) filed by Satish Yedage; and Dr. Monika Tripathi on April 13, 2026, for A Hybrid Multi-Model Cnn-Based System With Optimized Feature Extraction For Breast Cancer Detection And Classification.
Inventors include Satish Yedage; and Dr. Monika Tripathi.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: The present invention relates to a computer-implemented system and method for automated detection and classification of breast cancer using deep learning techniques. The invention utilizes a pretrained EfficientNetBO-based convolutional neural network integrated with a customized classification architecture to analyze mammogram images. The system performs preprocessing operations including image normalization, resizing to a fixed resolution, data augmentation, and class balancing to enhance input quality and model generalization. The processed images are fed into the EfficientNet-based feature extraction module, which captures hierarchical image features, followed by fully connected layers incorporating batch normalization, activation functions, and dropout for improved classification performance and reduced overfitting. The system enables end-to-end automated learning, eliminating the need for manual feature extraction. The model is trained and evaluated on large-scale mammography datasets and classifies images into multiple categories, including normal, benign, and malignant conditions. Performance is assessed using standard evaluation metrics such as accuracy, precision, recall, Fl-score, and ROC-AUC. Experimental results demonstrate that the proposed method achieves superior accuracy and reliability compared to conventional deep learning models. The invention provides a scalable, efficient, and accurate solution for early breast cancer detection, supporting clinical decision-making and improving diagnostic outcomes.
Disclaimer: Curated by HT Syndication.
