Artificial Intelligence: Based Approaches to Preclinical Pain Assessment

Monday, September 19th
8:30am12:00pm EDT

About the Session


  • Jennifer Lofgren, DVM, MS (USA)
  • Greg Neely, PhD, BSc (Australia)
  • Talmo Pereira, BSc, PhD (USA)
  • Guorong Wu, PhD (USA)
  • Daniel Pang, BVSc, PhD, DACVAA, DECVAA, MRCVS (Canada)
  • Scorrano Fabrizio, DVM, PhD (Switzerland)
  • Rick D’Eath, MA, DPhil (UK)
  • Duncan Lascelles, BSc, BVSc, PhD, FRCVS, CertVA, DSAS(ST) DECVS, DACVS (USA)

Measurement of pain in non-human species in laboratory, clinical and domesticated environments is limited to surrogate outcome measures due to their inability to communicate, unlike humans. In this context, the interpretation of the pain state is based on the measurement of behavior. The behavior of animals is dramatically influenced by the presence of humans, resulting in unpredictable biases during experiments.

Coupled with the increasing realization that the evaluation of complex behaviors may be more clinically relevant, has led researchers to explore a ‘hands-off’ methodology to measure complex behaviors. The explosive growth and availability of high-quality video and computational technology is facilitating the development of automated or semi-automated methods to capture and analyze complex behaviors as surrogate measures of pain and analgesia. Attendees will meet and interact with experts across a diverse range of non-human species (flies, rodents, companions, and farm animals) – experts who will explain, discuss and demonstrate the use of video and artificial intelligence (AI) in its application to the measurement of pain. AI-based automated video analysis is becoming increasingly available and utilized, and, as for any methodology, an awareness of its advantages and caveats will be essential for all scientists working with non-human species.