Mouse Behavior and Homeostatic Behaviors
Abstract
My research is focused on analyzing and quantifying feeding and drinking behaviors in mice to understand homeostatic regulation. In using computer vision and machine learning methods to to monitor mice, we create aa more naturalistic environment for observing mice, and enabling tracking of their posture and behavior. I am applying these insights to mouse models of diabetes and obesity.
Methodology
We use computer vision techniques and to track mice in long-term monitoring arenas. We then use supervised learning to create classifiers for behaviors we are interested in.
Key Findings
Our results show that our system can identify behavioral patterns and differences across different strains. These early analyses show interesting circadian insights.
Implications
This research provides tools and methodologies for studying homeostatic behaviors in mice, with potential applications in understanding feeding disorders, metabolic diseases, and basic behavioral neuroscience.