Real-time mastitis detection in livestock using deep learning and machine learning leveraging edge devices

Abstract

The livestock industry is a vital component of the global economy, with a value estimated at over $1.4 trillion. However, the health of livestock animals is frequently threatened by infectious diseases, which can have serious consequences for the industry and the economy. Bovine mastitis is one such disease that is prevalent and costly to treat. It is caused by bacterial infection of the mammary gland in cows and can have severe impacts on the dairy industry. In developing countries like Bangladesh, where the livestock sector is a significant contributor to the national economy, mastitis is a major concern. It is estimated that this disease costs the dairy industry millions of dollars each year in Bangladesh, due to reduced milk production, increased treatment costs, and culling of infected animals. The economic impact of mastitis can be particularly significant in a country like Bangladesh, where the livestock sector plays a vital role in the economy. In order to overcome this issue, this paper presents a real-time system for detecting mastitis in livestock using Deep Learning and Machine-Learning techniques leveraging edge devices. The proposed system aims to provide a timely and accurate diagnosis of clinical mastitis, ultimately reducing costs and improving the efficiency of treatment. By utilizing deep learning and machine learning techniques, the system is able to analyze data from edge devices and make accurate predictions about the presence of mastitis. This can help farmers and veterinarians identify infected animals and take appropriate action to prevent the spread of the disease. In the proposed system, various Deep Learning and Machine Learning algorithms were utilized for classification, and a comparison was made based on their accuracy and performance. The models that performed best with the highest accuracy were selected for further use. InceptionV3 and Random Forest algorithm were chosen for Deep Learning and Machine Learning, respectively, and had an accuracy of 99.34% and 99% respectively. A review of other papers that have used classification techniques for detecting mastitis shows that the models proposed in this paper have demonstrated better accuracy in the diagnosis of mastitis in livestock. The real-time system for detecting mastitis in livestock presented in this paper has the potential to significantly reduce the economic impact of this disease in the dairy industry of Bangladesh and other developing countries. By providing a timely and accurate diagnosis, the system can help to improve treatment efficiency and protect the health and productivity of livestock animals. In doing so, this system can have positive impacts on the livestock industry and the global economy by improving the health and productivity of livestock animals and reducing the costs associated with mastitis.