A machine-vision approach for automated pain measurement at millisecond timescales

Abstract

Objective and automatic measurement of pain in mice remains a barrier for discovery in both basic and translational neuroscience. Here we capture rapid paw kinematics during pain behavior in mice with high-speed videography and automated paw tracking with machine and deep learning approaches. Our statistical software platform, PAWS (Pain Assessment at Withdrawal Speeds), uses a univariate projection of paw position over time to automatically quantify fast paw dynamics at the onset of paw withdrawal and also lingering pain-related behaviors such as paw guarding and shaking. Applied to innocuous and noxious stimuli across six inbred mouse strains, a linear discriminant analysis reveals a two-dimensional subspace that separates painful from non-painful stimuli on one axis, and further distinguishes the severity of pain on the second axis. Automated paw tracking combined with PAWS reveals behaviorally-divergent mouse strains that display hypo- and hyper-sensitivity to mechanical stimuli. To demonstrate the efficacy of PAWS for detecting hypersensitivity to noxious stimuli, we chemogenetically activated pain-aversion neurons in the amygdala, which further separated the behavioral representation of pain-related behaviors along a low-dimensional path. Taken together, this automated pain quantification approach should increase the ease and objectivity of collecting rigorous behavioral data, and it is compatible with other neural circuit dissection tools for determining the mouse pain state.