MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541124252 A) filed by Koneru Lakshmaiah Education Foundation, Hyderabad, Telangana, on Dec. 9, 2025, for 'deep models based skin disease classification for efficient cancer detection.'
Inventor(s) include Lakshmi Deepthi Nukala; and Dr. T. Aruna Sri.
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 computer-implemented deep learning framework for automated classification of skin diseases, including melanoma, through multi-model convolutional architectures. The system integrates Convolutional Neural Networks (CNN), VGG16, DenseNet, and ResNet models to analyze dermoscopic images obtained through ROI-based preprocessing. The architecture extracts hierarchical features using stacked convolution, pooling, and dense layers, achieving high accuracy and reduced loss across multiple epochs. Using ISIC dermoscopic datasets comprising seven disease classes, the invention demonstrates significantly improved diagnostic performance compared to conventional image processing approaches such as GLCM, HE, and Gabor filtering. Benchmark evaluations show that DenseNet and ResNet deliver superior accuracy due to deeper connectivity, efficient gradient flow, and residual learning mechanisms. The invention further includes comparative accuracy-loss graphs and architectural diagrams to illustrate model performance. The disclosed system provides an efficient, scalable, and automated diagnostic tool for early cancer detection, reducing diagnostic delays and improving clinical outcomes."
Disclaimer: Curated by HT Syndication.
