Artificial intelligence and laboratory biomarkers in veterinary medicine: an update about machine learning applications.

Abstract

The use of artificial intelligence has been a revolution in human medicine in the last years. Machine learning (ML), a subset of artificial intelligence, has allowed the use of a large amount of data to search for associations of clinical significance. Although the use of ML with laboratory biomarkers in veterinary medicine is limited in comparison with the human side, the number of studies published in this field is increasing in recent years. This review addresses some basic concepts and gives a general overview of the use of ML in the area of biomarkers in veterinary medicine. It is especially focused on two species of each companion (dogs and cats) and farm (bovine and porcine) animals, and provides information about the advances made in the last years, indicating how ML using laboratory data can contribute to the diagnosis and monitoring of selected diseases and, in case of farm animals, also to better control and improve their productive performance. In addition, possible new applications such as optimizing laboratory management, refining diagnostics or improving treatment monitoring based on the experience in humans are outlined. This narrative review can contribute to a better knowledge about the possible use of ML in laboratory biomarkers in veterinary and animal sciences for current and future applications.