Machine learning fostering the judicious use of antimicrobials in bovine mastitis control

Mastitis is the most prevalent and costly disease in dairy production and a leading driver of antimicrobial use (AMU) on farms. Given the global public health threat posed by antimicrobial resistance (AMR) and increasing pressure to limit AMU in animal production, innovative tools to promote antimicrobial stewardship in mastitis control are urgently needed. Machine learning (ML) models offer promising capabilities by enhancing diagnostics, improving the accuracy of selective therapy criteria, and guiding treatment and prevention decisions, thereby contributing to AMR mitigation. ML applications in mastitis control have the potential to simplify, standardize, and optimize on-farm management practices. This short review provides an overview of the potential uses of ML in mastitis control, focusing on promoting judicious AMU through: a) mastitis diagnosis using automated milking system data, b) supporting decision-making in selective therapy protocols, and c) predicting AMR in mastitis pathogens using mass spectrometry data.