Using machine learning to automate the analysis of an olfactory habituation-dishabituation task in mice

IntroductionImproving the efficiency and accuracy of annotation and extraction of performance data from mouse behavioural tasks will improve both the throughput and scientific value of preclinical research. MethodsHere, we present and validate an automated pipeline for the annotation and quantification of performance in a mouse olfactory habituation-dishabituation task, using a single side-view camera, resulting in occluded body parts. We created a pipeline for task analysis, combining DeepLabCut, for pose-estimation, and SimBA, for behavioural classification to automatically quantify odour interaction (sniffing time) in a three-odour (water, familiar mouse social odour, novel mouse social odour) variant of the task. We used a subset of previously published, fully manually annotated datasets to train the models and unseen videos from the same study to validate the utility of our machine learning pipeline. Results and conclusionOur analysis pipeline estimated behavioural performance in the task with high accuracy, and the data produces similar technical and biological results to manual methods when analysed by linear mixed modelling. Thus, we validated the utility of our new pipeline for the automated scoring of this mouse sensory task.