MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541124380 A) filed by Vishnu Institute Of Technology, Kovvada, Andhra Pradesh, on Dec. 10, 2025, for 'hybrid cnn-transformer architecture for enhanced skin cancer classification.'
Inventor(s) include Abdul Rahman Shaik; and K. Kiran.
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: "The invention discloses a hybrid deep-learning architecture that integrates Convolutional Neural Networks (CNNs) with Transformer-based self-attention mechanisms for accurate classification of skin cancer from dermoscopic images. The system includes preprocessing operations such as resizing, normalization, augmentation, and class balancing, followed by CNN blocks that extract local lesion features. A Transformer layer applies multi-head self-attention to model global contextual relationships, producing a comprehensive feature representation. A fully connected softmax classifier generates lesion category predictions with high diagnostic accuracy. Through optimized hyperparameters, dropout, and batch normalization, the model achieves robust generalization, outperforming traditional CNN approaches and achieving approximately 97% accuracy on benchmark datasets. The invention supports deployment in clinical, mobile, and telemedicine diagnostic workflows."
Disclaimer: Curated by HT Syndication.
