SheepEye: a based-web app for real-time diagnosis of sheep anemia

Abstract

Haemonchus contortus is an extremely harmful blood-feeding nematode affecting small ruminants, leading to anemia, weight loss, and, in severe cases, animal death. Traditional methods of monitoring anemia in sheep, such as regular physical examinations by veterinarians and laboratory tests, can be expensive and time-consuming. In this work, we propose an anemia monitoring system that uses a web-based app. The methodology for the SheepEye app is based on deep learning algorithms, including the U-net model for segmentation and the VGG19 model for classification. All learning algorithms, as well as the development of the app, were implemented in Python. The SheepEye web-based app is a promising technology that can facilitate and improve the diagnosis of parasitic infections in sheep and enhance sheep productivity. By using the app, farmers can detect anemia in their flocks and implement target selective treatment, which reduces the use of anthelmintics and consequently minimizes the risk of parasitic resistance. The SheepEye app is still in a prototype stage, but its prospective is extremely promising and our goal is to further develop it so that it can be made available for producers to use.