Artificial Intelligence and Animal Health

Context and issues

Artificial intelligence (AI) is a broad set of theories and technologies used to solve problems of high logical or algorithmic complexity. In a broad sense, AI is used in animal health to address complex issues such as quantitative and predictive epidemiology or precision animal and human medicine, and also to study host-pathogen interactions in detail. However, AI methods are still not well known and mobilised in their diversity in the field of health, particularly veterinary health. Yet, AI addresses challenges that make sense in animal health: better understanding a situation and its dynamics, such as an epidemic dynamic, better perceiving the environment, for example by automatically detecting patterns, shapes or signals at different scales, or assisting human decision-making, for example through expert systems or diagnostic assistance. AI offers a wide range of methods for this purpose, including learning methods, but also methods for solving complex problems, automating tasks or reasoning, integrating heterogeneous information, or providing decision support.

Results

On the basis of a literature review of scientific articles at the interface between AI and animal health, covering the period from 2009 to 2019, and via interviews with French researchers positioned at this interface, the main areas of animal health where AI approaches are currently being mobilised were identified. We have analysed how AI can contribute to renewing research issues in animal health and remove methodological or conceptual barriers. Thus, AI contributes to the diagnosis and detection of diseases, to making predictions more reliable and reducing interpretation errors, to producing more realistic representations of biological systems, to increasing the readability of computer codes for their users, to speeding up decisions and improving the accuracy of risk analyses, and finally to better targeting interventions and anticipating possible negative effects. In turn, animal health challenges can stimulate AI research because of the specificity of the systems, data, constraints and objectives. After presenting the possible obstacles and levers to the use of AI in animal health, we have formulated recommendations to better understand the challenge of this interface between formal and life sciences disciplines.

Perspectives

With the development of several recent concepts promoting a global and multisectoral perspective of health, AI should contribute to defragmenting the different health disciplines towards a more transversal and integrative research.

Bibliographical references

Ezanno P., Picault S., Beaunée G., Bailly X., Munoz F., Duboz R., Monod H., Guégan J-F. 2021. Research perspectives on animal health in the era of artificial intelligence. Veterinary Research 52, 40, https://doi.org/10.1186/s13567-021-00902-4

Positionnement de quelques thèmes de recherche en santé animale (en noir) et de méthodes d’IA utiles pour les développer (en vert). Auteure : P. Ezanno.