MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541125459 A) filed by Dharsana Dharani VG; Dharanika S; Ramya Bharathi K; and Dr. E. Uma, Chennai, Tamil Nadu, on Dec. 11, 2025, for 'an advanced efficientnetv2-mamba hybrid framework for multi-class classification in dermoscopic imaging.'
Inventor(s) include Dharsana Dharani VG; Dharanika S; Ramya Bharathi K; and Dr. E. Uma.
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: "Accurate identification of skin diseases is crucial for timely diagnosis and effective treatment. This study proposes a hybrid deep learning architecture that integrates the strong feature extraction capability of EfficientNetV2 with the sequence-processing strength of the Mamba State Space Model (SSM). The work focuses on three major categories of skin disorders-Melanoma, Non-Melanoma Skin Cancer, and Inflammatory/Benign conditions using a dataset of 8,655 dermoscopic images. Multiple benchmark models, including ResNet-50, EfficientNet-B0, Vision Transformer, standalone Mamba, and the proposed EfficientNetV2-Mamba hybrid system, are evaluated for comparison. In the proposed framework, EfficientNetV2 generates rich hierarchical feature representations which are converted into sequential tokens and passed to Mamba to capture long-range spatial relationships. The resulting sequence embeddings are aggregated using global average pooling and subsequently classified through a fully connected layer."
Disclaimer: Curated by HT Syndication.
