Published Paper Details:

This survey aims to automate the process of identification and diagnosis of Haemoprotozoan diseases in cattle, using machine learning algorithms to build a model from a dataset of images and diagnoses collected from local labs. The project focuses on three significant Haemoprotozoan diseases that pose a significant risk to cattle all across India: Anaplasmosis, Babesiosis, and Theileriosis. The model is designed to run on a client-server web application, where a picture of the bloodstream sample is sent to the server, which then returns two parameters specifying whether the diagnosis is positive or negative, and the type of disease. This could significantly reduce mortality rates by making the process cheaper and faster. The model will be tested and evaluated several times in the training and testing modules. The evaluation process will be based on the model's accuracy, as well as other factors such as the model's speed and cost-effectiveness. The outcomes are then differentiated from the existing models already in use, to inspect the accuracy and implementation of the model. The final model will be deployed in the client-server web application.