MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641052026 A) filed by Dasari Divya; and St. Peter's Engineering College, Hyderabad, Telangana, on April 23, for 'an optimized convolutional neural networks and deep learning methods in a hybrid xception-inceptionv3 model for skin disorder diagnosis.'
Inventor(s) include Mr. Ch V Ganesh; Mr. Voruganti Santhosh Kumar; Mrs. Vempati Vineetha; Mr. D Narasimha Rao; Mr. Pandurangachary; Mrs. P Amaleswari; Mr. Poola Mallikarjun; Mr. Katru Madhu; and Ms. Lallu Sekharan.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "This invention presents a hybrid Xception-InceptionV3 structure for diagnosing multiple skin disorders by leveraging the complementary strengths of two superior deep learning models. In medical image classification tasks, deep learning (DL), particularly convolutional neural networks (CNNs), has demonstrated remarkable effectiveness. This study presents a hybrid DL method that combines the Xception (XCP) and InceptionV3 (INCP3) models to categorize various kinds of skin diseases. This invention makes use of data augmentation, class balancing, and transfer learning to reduce dataset imbalance, improve predictability. Adam optimizer, dropout regularization used to prevent overfitting. With a training accuracy of 99.01and a validation accuracy of 89.56%, experimental evaluation on benchmark datasets, like HAM10000, demonstrated good performance. Additional metrics such as precision, recall, F1-score, confusion matrices, ROC analysis further confirmed robustness of the approach in distinguishing wide range skin disease categories. These results highlight the usefulness of using hybrid DL models in computer-assisted skin cancer diagnosis."
Disclaimer: Curated by HT Syndication.
