Related Experiment Video
Updated: May 2, 2026
08:38
A System for Tracking the Dynamics of Social Preference Behavior in Small Rodents
Published on: November 21, 2019
7.0K
RMScratcher: An edge-AI platform for high-throughput analysis of rodent scratching behavior
Moran Zhang1, Xin Zeng2, Min Chen1
1Department of Biomedical Engineering, Southern University of Science and Technology, Shenzhen, Guangdong 518055, China.
Journal of Neuroscience Methods
|April 30, 2026
Summary
RMScratcher accurately quantifies rodent scratching behavior, including the scratching region, using edge computing. This platform enhances behavioral neuroscience research for somatosensory and neuroimmune studies.
Area of Science:
- Behavioral Neuroscience
- Computational Neuroscience
- Neuroimmunology
Background:
- Quantitative analysis of rodent scratching is crucial for understanding somatosensory and neuroimmune mechanisms.
- Scalable and high-throughput analysis of rodent scratching behavior presents a significant challenge.
Purpose of the Study:
- To develop an automated, high-throughput platform for accurate quantitative analysis of rodent scratching behavior.
- To enable precise identification of scratching bouts, touch, duration, and the specific scratching region.
Main Methods:
- Development of RMScratcher, a compact platform integrating multi-channel video acquisition and Jetson TX2-based edge computing.
- Utilized an enhanced YOLOv8s framework with GhostConv and Context Aggregation modules for efficient and accurate behavior detection.
- Minimal hardware requirements for implementation.
Main Results:
- RMScratcher achieved over 95% accuracy in distinguishing scratching from other rodent behaviors.
- Reliable quantification of scratching bout, touch, duration, and identification of the scratching region.
- Robust performance demonstrated across diverse rodent models and experimental conditions.
Conclusions:
- RMScratcher significantly advances the methodological toolkit for behavioral neuroscience research.
- The platform integrates video recording with intelligent computation for reproducible studies.
- Facilitates research in somatosensory function, neuroimmune interactions, behavioral phenotyping, and abnormal sensory behaviors.
