MUMBAI, India, April 17 -- Intellectual Property India has published a patent application (202621016169 A) filed by Nivedita Shimbre; and Dr. Pushpalata Aher, Nashik, Maharashtra, on Feb. 13, for 'a multimodal skin melanoma diagnostic system based on hybrid firefly-ant colony optimization and deep cnn-bilstm model.'
Inventor(s) include Nivedita Shimbre; and Dr. Pushpalata Aher.
The application for the patent was published on April 17, under issue no. 16/2026.
According to the abstract released by the Intellectual Property India: "The present invention discloses a multimodal skin melanoma diagnostic system based on hybrid Firefly-Ant Colony Optimization and deep CNN-BiLSTM architecture for automated and accurate melanoma detection. The system integrates dermoscopic skin lesion images and corresponding clinical metadata to enhance diagnostic reliability. Image preprocessing and metadata preprocessing modules perform data normalization, noise removal, contrast enhancement, missing value imputation, categorical encoding, and standardization. A hybrid Firefly-Ant Colony Optimization algorithm is employed for intelligent feature selection and dimensionality reduction. The optimized features are processed through a dual-branch deep learning architecture comprising a CNN branch for spatial feature extraction and a BiLSTM branch for bidirectional metadata sequence modeling. A multimodal feature fusion layer integrates both feature representations into a unified feature vector. A softmax classifier performs multiclass skin lesion classification and generates melanoma and non-melanoma predictions along with confidence scores. The system provides a scalable, real-time, and clinically deployable solution for early-stage melanoma detection and decision support."
Disclaimer: Curated by HT Syndication.
