@article{fdi:010079727,
  title = {{I}ntelligence artificielle et sant{\'e} animale},
  author = {{E}zanno, {P}. and {P}icault, {S}. and {W}inter, {N}. and {B}eaunee, {G}. and {M}onod, {H}. and {G}u{\'e}gan, {J}ean-{F}ran{\c{c}}ois},
  editor = {},
  language = {{FRE}},
  abstract = {{M}obilizing {A}rtificial {I}ntelligence ({AI}) approaches in {A}nimal {H}ealth ({AH}) makes it possible to address issues of high logical or algorithmic complexity such as those encountered in quantitative and predictive epidemiology, precision-based medicine, or to study host x pathogen relationships. {AI} can to some extent facilitate diagnosis and case detection, make predictions more reliable and reduce errors, allow more realistic representations of complex biological systems also readable by non-computer scientists, speed-up decisions, improve accuracy in risk analyses, and allow interventions to be better targeted and their effects anticipated. {I}n addition, challenges in {AH} may stimulate {AI} research in turn due to the specificity of systems, data, constraints, and analytical objectives. {B}ased on a literature review at the interface between {AI} and {AH} covering the period 2 0092019, and interviews with {F}rench researchers positioned at this interface, this synthesis explains the main areas in {AH} in which {AI} is mobilized, how it contributes to revisiting {AH} research issues and removes methodological barriers, and how {AH} research questions stimulate new {AI} research development. {A}fter presenting the possible obstacles and levers, we propose recommendations to better grasp the challenge represented by this new {AH}/{AI} interface.},
  keywords = {},
  booktitle = {},
  journal = {{INRA} {P}roductions {A}nimales},
  volume = {33},
  numero = {2},
  pages = {95--108},
  ISSN = {2273-774{X}},
  year = {2020},
  DOI = {10.20870/productions-animales.2020.33.2.3572},
  URL = {https://www.documentation.ird.fr/hor/fdi:010079727},
}