Inteligência artificial na medicina veterinária: fundamentos teóricos, aplicações, desafios e perspectivas

Abstract

In 1956, Marvin Minsky and John McCarthy coined the term "artificial intelligence" (AI) to refer to "the science and engineering of creating intelligent machines." Since then, the field of AI has grown and become relevant. AI has also gained ground in veterinary medicine, such that the way the profession is practiced is being reshaped by this technology. Thus, the objective of this work is to conduct a literature review with a qualitative and exploratory approach on artificial intelligence in veterinary medicine: theoretical foundations, applications, challenges, and perspectives. To this end, a bibliographic search was conducted in the IEEEXplore, Web of Science, Google Scholar, PubMed, and Scopus databases; and the terms "artificial intelligence and veterinary medicine," "artificial intelligence and animal health," "machine learning and animal health," and "machine learning and veterinary medicine" were used to locate works on AI and veterinary medicine. From the search results, works published between 2022 and 2025 were selected in order to obtain a current and relevant literature review on aspects of AI in veterinary medicine. The theoretical foundations of AI involve machine learning, as through this process, AI analyzes data using learning algorithms and solves problems. This technology is applicable in many fields within veterinary medicine, from diagnosis and clinical support, in areas such as diagnostic imaging, assisting in the interpretation of ultrasounds, magnetic resonance imaging, and other imaging exams, to remote animal monitoring, clinical decision support through AIs trained with scientific literature and clinical cases; animal pathology; predictive analytics and epidemiology, also contributing to one health; research and development; workflow automation, reducing stress from overload through the automation of repetitive tasks; and education, through customized GPTs, among other functionalities. However, some challenges still need to be faced, limitations such as the lack of quality data, the need for trained veterinary professionals; financial resources; barriers to the ethical, responsible, and transparent use of AI; and the absence of regulation can affect the performance of this technology. Despite these factors, it is expected that AI will continue to revolutionize veterinary medicine through efficiency, accuracy, and speed in performing tasks. In this scenario, AI should not replace the veterinarian, but be a support tool for the professional. Finally, it is concluded that AI has become relevant to veterinary medicine and can be applied in various areas. Retrospective studies that gather large amounts of data and research on machine learning can be useful for training AI models geared towards veterinary medicine. Furthermore, research on transparent software can also contribute to the ethical and responsible use of this technology.