MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641041624 A) filed by Dayananda Sagar University, Bengaluru, Karnataka, on April 1, for 'dermaai: ai-powered skin disease detection severity analysis & treatment recommendation system.'

Inventor(s) include Sushma DS; A Ranjini; Seema J Kampli; Megha HN; Mamatha R M; Manish Nandy; Giridharan KS; Gudivada Pardha Srikanth; Gowtham KA; and Sabareesh SP.

The application for the patent was published on May 29, under issue no. 22/2026.

According to the abstract released by the Intellectual Property India: "Skin diseases represent a significant global health burden, requiring early and accurate diagnosis for effective treatment. DERMAAI, an Al-based system is designed for automated skin disease detection, severity assessment, and personalized treatment recommendation. The framework integrates deep learning, image processing, and clinical decision-support into a unified pipeline. For detection, a CNN based on EfficientNet-:B4 with transfer learning extracts features from dermoscopic and clinical images. Preprocessing includes artifact removal using the DullRazor algorithm, illumination correction, and lesion segmentation using a U-Net model. To address class imbalance, SMOTE and data augmentation are applied. For severity assessment, deep features are combined with key clinical characteristics such as lesion shape, border, color, and texture. Additional features from GLCM-based texture analysis, contour detection, and color histograms are fused and analyzed using XGBoost to classify severity into mild, moderate, or severe. The treatment recommendation module uses a knowledge-based system with BERT-based natural language processing to map predicted conditions and severity to appropriate treatments. Grad-CAM further enhances interpretability by highlighting important image regions. The proposed DERMAAI system offers a scalable, interpretable, and efficient solution for real-time skin disease diagnosis and management, particularly in telemedicine and resource-limited settings."

Disclaimer: Curated by HT Syndication.