Abstract
Skeletal movement is essential for a wide range of behaviors, including locomotion, feeding, and grooming. However, quantifying skeletal kinematics in freely moving rodents has been challenging due to the small size and complex anatomy of these animals. Here, we present a new tracking method that uses high-speed videography and machine learning to track the movement of individual bones in freely moving rodents. Our method is able to track the movement of bones with high accuracy and precision, and can be used to study a wide range of behaviors. We demonstrate the utility of our method by tracking the movement of the forelimb and hindlimb bones during locomotion. Our results provide new insights into the kinematics of rodent locomotion, and highlight the potential of our method for studying skeletal movement in a variety of other behaviors.
Introduction
Skeletal movement is essential for a wide range of behaviors, including locomotion, feeding, and grooming. Understanding the kinematics of skeletal movement is therefore critical for understanding how rodents interact with their environment. However, quantifying skeletal kinematics in freely moving rodents has been challenging due to the small size and complex anatomy of these animals.
Traditional methods for tracking skeletal movement, such as motion capture and videography, are often limited by their invasiveness or their inability to track the movement of individual bones. More recently, machine learning-based methods have been developed that can track the movement of bones from high-speed videography. However, these methods have typically been limited to tracking the movement of a small number of bones, and have not been validated for use in freely moving rodents.
Here, we present a new tracking method that uses high-speed videography and machine learning to track the movement of individual bones in freely moving rodents. Our method is able to track the movement of bones with high accuracy and precision, and can be used to study a wide range of behaviors. We demonstrate the utility of our method by tracking the movement of the forelimb and hindlimb bones during locomotion. Our results provide new insights into the kinematics of rodent locomotion, and highlight the potential of our method for studying skeletal movement in a variety of other behaviors.
Methods
Animals
All experiments were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee of the University of California, Berkeley. Male C57BL/6J mice (8-12 weeks old) were used in all experiments.
Experimental setup
Mice were placed in a custom-built arena (30 cm x 30 cm x 30 cm) made of clear Plexiglas. The arena was placed on a vibration isolation table to minimize movement artifacts. High-speed videography was performed using a Photron Fastcam SA3 camera (Photron, Tokyo, Japan) operating at 500 frames per second. The camera was positioned above the arena and focused on the mouse's body.
Tracking method
Our tracking method is based on a deep learning algorithm called
