Intellectual Property India Publishes Patent Application for Hybrid Deep Learning And Machine Learning Approach For Skin Cancer Classification Using Dermoscopic Images

MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641069921 A) filed by Sasi Institute Of Technology & Engineering; M. Pavani; M Satya Srinivas; S Siva Ramaraja; and V Srinivas on June 04, 2026, for Hybrid Deep Learning And Machine Learning Approach For Skin Cancer Classification Using Dermoscopic Images.

Inventors include Sasi Institute Of Technology & Engineering; M. Pavani; M Satya Srinivas; S Siva Ramaraja; and V Srinivas.

The application for the patent was published on June 12, 2026, under issue no. 24/2026.

Abstract: Skin cancer is one of the most common and dangerous forms of cancer worldwide, where early detection plays a vital role in improving patient survival rates. Manual diagnosis of dermoscopic images is often challenging due to the visual similarity between different skin lesion categories. To address this issue, the proposed work presents a hybrid deep learning and machine learning framework for automated skin cancer classification using dermoscopic images. The proposed system utilizes the EfficientNetB0 deep learning model for extracting discriminative visual features from skin lesion images. The extracted features are optimized using Principal Component Analysis (PCA) to reduce dimensionality and improve computational efficiency. To handle dataset imbalance, the Synthetic Minority Oversampling Technique (SMOTE) is applied during model training. The optimized feature vectors are then classified using the XGBoost machine learning algorithm for multi-class skin lesion classification. The framework is trained and evaluated using the HAM10000 dataset, which contains seven categories of dermoscopic skin lesions including melanoma, basal cell carcinoma, benign keratosis, dermatofibroma, vascular lesions, actinic keratosis, and melanocytic nevus. In addition, a Flask-based web application is developed to enable users to upload dermoscopic images and obtain prediction results along with confidence scores in real time. Experimental results demonstrate that the proposed hybrid approach achieves classification accuracy of approximately 78–80%, showing the effectiveness of combining deep learning and machine learning techniques for automated skin cancer diagnosis. The proposed system can assist dermatologists in early-stage skin cancer detection and support improved clinical decision-making.

Disclaimer: Curated by HT Syndication.