Towards a fully automated surveillance of well-being status in laboratory mice using deep learning: Starting with facial expression analysis

Abstract

Assessing the well-being of an animal is hindered by the limitations of efficient communication between humans and animals. Instead of direct communication, a variety of parameters are employed to evaluate the well-being of an animal. Especially in the field of biomedical research, scientifically sound tools to assess pain, suffering, and distress for experimental animals are highly demanded due to ethical and legal reasons. For mice, the most commonly used laboratory animals, a valuable tool is the Mouse Grimace Scale (MGS), a coding system for facial expressions of pain in mice. We aim to develop a fully automated system for the surveillance of post-surgical and post-anesthetic effects in mice. Our work introduces a semi-automated pipeline as a first step towards this goal. A new data set of images of black-furred laboratory mice that were moving freely is used and provided. Images were obtained after anesthesia (with isoflurane or ketamine/xylazine combination) and surgery (castration). We deploy two pre-trained state of the art deep convolutional neural network (CNN) architectures (ResNet50 and InceptionV3) and compare to a third CNN architecture without pre-training. Depending on the particular treatment, we achieve an accuracy of up to 99% for the recognition of the absence or presence of post-surgical and/or post-anesthetic effects on the facial expression.

Reliability of the Mouse Grimace Scale in C57BL/6JRj Mice

“…To further improve the reliability of MGS, the issues raised need to be addressed. A further approach to generate more objective and reliable data is to standardize the experimental set-up for MGS scoring, including image/video generation, training, as well as scoring sessions, and to automatize the facial expression analysis of mice [40][41][42].…”

The Utility of Grimace Scales for Practical Pain Assessment in Laboratory Animals

“…Their system was highly accurate (94% agreement with human scores) for a binary (pain versus no pain) output, with scores correlating highly with human-assigned scores. Other groups have similarly demonstrated the promise of deep learning methods for use with the MGS when based on binary outputs [ 100 , 101 ]. Progress has also been made toward automating a facial pain expression system in sheep using techniques used in human facial recognition [ 102 , 103 ].…”

Grimace Scores: Tools to Support the Identification of Pain in Mammals Used in Research

“…A future area for development and benefit is the use of software automation in the development and scoring of facial expressions. The use of scoring software along with the installation of video cameras into enclosures may be able to enhance and hasten the development of grimaces, offer highly accurate grimace scores for animals in pain but also allow the remote monitoring and scoring of affected animals [ 41 , 59 , 134 , 135 ].…”

“…A standard training program would be useful for grimace score users and has been useful for other pain scoring systems [ 38 , 46 , 86 ]. Part of the development, training and implementation of grimaces could be enhanced by the use of various technologies such as automated or semiautomated software for scale development and scoring via video surveillance [ 41 , 59 , 134 , 135 ]. These nascent technologies are often unfeasible due to cost, infrastructure constraints and a lack of development but in the future their use may play a greater role in grimace scoring systems.…”