MUMBAI, India, Oct. 26 -- Intellectual Property India has published a patent application (202611107713 A) filed by Maharishi Markandeshwar Deemed To Be University on September 05, 2026, for An Integrated Deep Learning And Machine Learning Ensemble Framework For Precise Prediction Of Lumpy Skin Disease.
Inventors include Ritu Aggarwal; and Eshaan Aggarwal.
The application for the patent was published on October 23, 2026, under issue no. 43/2026.
Abstract: An integrated deep learning framework utilizing a pre-trained VGG19 convolutional neural network is disclosed for the precise and early prediction of lumpy skin disease (LSD) in cattle and buffaloes. Trained on a comprehensive dataset of 24,804 high-resolution image samples (comprising 12,000 healthy and 12,804 LSD- infected cattle), the system processes input images preprocessed to a standardized 256×256 resolution with data augmentation techniques including multi-angle rotations, flipping, and cropping. Utilizing transfer learning fine-tuned for veterinary diagnostics, the VGG19 architecture extracts spatial features to compute probability distributions via ReLU and softmax activation layers. The proposed ensemble model outperforms comparative deep learning classifiers (including Xception, ResNet152V2, InceptionV3, and DenseNet121), achieving a classification accuracy of 98.71%, an Area Under Curve (AUC) of 0.9940, a precision of 0.9418, an F1- score of 0.9149, and an execution time of 1.18 seconds. The framework enables rapid, non-invasive, and cost-effective field surveillance to mitigate viral disease transmission.
Disclaimer: Curated by HT Syndication.
