A machine learning tool with light-based image analysis for automatic classification of 3D pain behaviors.

Abstract

Detailed assessment of pain-related behaviors in animals is essential for both exploring pain mechanisms and evaluating analgesic efficacy. While pose estimation tools have advanced automated behavior analysis, current existing algorithms often do not account for an animal's body-part contact intensity with-and distance from-the surface, a critical nuance for measuring certain pain-related responses like paw withdrawal ("flinching"). These subtle responses continue to require time-consuming and subjective human scoring. Here, we present BAREfoot (behavior with automatic recognition and evaluation), a supervised machine learning (ML) algorithm that combines pose estimation with light-based analysis of body-part contact and elevation to automatically detect pain behaviors in freely moving mice. We show the utility and accuracy of this algorithm for capturing a range of pain-related behavioral bouts using a bottom-up animal behavior platform and its application for robust drug screening. This open-source algorithm is adaptable for detecting diverse behaviors across species and experimental platforms.