Automated detection of social avoidance in mice using machine learning-based analysis through kinematic segmentation

초록

Accurately assessing social behavior in rodent models is crucial for advancing research into social dysfunction and related neurobiological mechanisms. Asocial behaviors, such as social avoidance and disengagement, are challenging to measure objectively due to their subtle and transient nature, which traditional observer-based methods struggle to capture without introducing bias. To address these challenges, we developed a machine learning-based tool that uses kinematic segmentation to detect antisocial behaviors in mice. By analyzing movement patterns and interaction dynamics, this tool provides objective and reproducible measurements of social avoidance behavior, uncovering behavioral nuances that may remain hidden with conventional methods. This approach not only enhances precision in behavioral assessment but also offers a deeper understanding of antisocial tendencies in rodent models.