This project aims to develop an image-based deep learning model for the early and accurate detection of Lumpy Skin Disease Virus (LSDV) in cattle. Given the challenges associated with traditional diagnostic methods, which require extensive resources and specialized expertise, our proposed approach will focus on creating a lightweight, efficient convolutional neural network (CNN) model that can be easily deployed in rural and resource-limited settings. Building on existing research, we will leverage CNN architectures such as MobileNetV2 and ResNet50, which are known for their suitability in mobile and real-time applications, enabling remote diagnostics and fast decision-making.
Lumpy Skin Disease Virus (LSDV) is a significant threat to the cattle industry, causing substantial economic losses due to decreased milk production, weight loss, and mortality in affected animals . Traditional diagnostic methods rely on clinical examination and laboratory tests, which can be time-consuming and challenging to deploy in resource-limited settings, such as rural farms . The growing availability of image-based diagnostic tools, along with the advancements in deep learning algorithms, offers an innovative alternative for the rapid and accurate detection of animal diseases, including LSDV .
• Develop a deep learning model for LSDV detection: Create a model capable of accurately identifying the presence of LSDV in cattle images, focusing on lesions and skin symptoms. • Enhance accessibility for rural veterinarians: Design the diagnostic tool to work with minimal technical resources, enabling veterinarians to use it in the field. • Validate the model’s accuracy and reliability: Conduct testing in various field conditions to ensure robustness and adaptability. • Build an alert and reporting system: Include a feature to notify relevant authorities in case of detected outbreaks to ensure quick response and containment.