MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202641009052 A) filed by Koneru Lakshmaiah Education Foundation; Chinaguravaiah Makkena; Alampally Sreedevi; Hari Prasada Raju Kunadharaju; Ramakrishna Akella; and Sai Krishna Gaduputi Subbammagari, Hyderabad, Telangana, on Jan. 29, for 'a shiftwise convolution-based deep learning system and method for automated multi-class skin lesion classification using dermoscopic images.'
Inventor(s) include Chinaguravaiah Makkena; Alampally Sreedevi; Hari Prasada Raju Kunadharaju; Ramakrishna Akella; and Sai Krishna Gaduputi Subbammagari.
The application for the patent was published on Feb. 13, under issue no. 07/2026.
According to the abstract released by the Intellectual Property India: "The present invention discloses a Shiftwise Convolution-based deep learning system and method (SW-DermNet) for automated multi-class skin lesion classification using dermoscopic images. The proposed system effectively captures both local texture details and large contextual lesion patterns while maintaining computational efficiency. Experimental evaluation on HAM10000 and ISIC datasets demonstrates an overall classification accuracy of approximately 93%, melanoma sensitivity of up to 88-89%, specificity of about 95%, and an AUC of approximately 0.97, thereby outperforming existing deep learning models. The invention provides a robust and clinically applicable solution for early skin cancer detection."
Disclaimer: Curated by HT Syndication.
