A review on the impact of AI-enabled thermal imaging and IoT sensor fusion on early detection of mastitis in dairy cattle

One of the most emerging economic diseases of dairy cattle, mastitis, reduces animal welfare, produces costly veterinary bills, prevents the animal from reaching its full milk production potential, creates the need to cull cows from the herd, and reduces the overall efficiency of the farm system. Conventional detection methods such as clinical inspection, the California Mastitis Test (CMT), and somatic cell (SC) functions, while useful, can identify subclinical cases at such an early enough stage for intervention. New technologies are turning towards the development of mastitis detection in real time and remotely using IoT devices, software, and complex AI algorithms. AI-driven analytics, coupled with infrared thermal imaging IRT and state of the art diagnostic IoT devices such as milking systems and behavioural collars, have yet to be integrated. IRT devices have the capability to analyse udder temperature and map inflammation. AI algorithms on image classification or data fusion from single tool-based approaches are able to achieve precision above the rest. Market and technological readiness, alongside cost and environmental variability are still fuelling the debate on the practical use of the technologies. This study slams breakthroughs and obstacles on record for detection of mastitis with the help of AI, Image fusion technologies, integrated IRT sensors, and IoT systems, while the rest argue the integration methods for cost-effective, vigilant husbandry of dairy cows in the herd. • This review study slams breakthroughs and obstacles on record for detection of mastitis with the help of AI, Image fusion technologies, integrated IRT sensors, and IoT systems. • This review study argues the integration methods for cost-effective, vigilant husbandry of dairy cows in the herd. • The integration of IoT devices, sensors, platforms, and analytics improves the reliability and trustworthiness of animal health monitoring in farms.