Related Experiment Video
Updated: Jun 7, 2025
07:33
Non-invasive Skeletal Muscle Quantification in Small Animals Using Micro-computed Tomography
Published on: November 8, 2024
345
Highly accurate and precise determination of mouse mass using computer vision.
Malachy Guzman1,2, Brian Q Geuther1, Gautam S Sabnis1
1The Jackson Laboratory, Bar Harbor, ME, USA.
Patterns (New York, N.Y.)
|November 21, 2024
Summary
Researchers developed a non-invasive method using computer vision to estimate mouse body mass from videos. This approach reduces animal stress and allows for continuous monitoring, improving preclinical research and animal welfare.
Area of Science:
- Biomedical Engineering
- Animal Science
- Computer Vision
Background:
- Body mass changes are critical health indicators in animal studies.
- Current manual weighing methods for rodents induce stress and provide static data.
- A non-invasive, continuous mass monitoring system is needed for preclinical research.
Purpose of the Study:
- To develop and validate a computer vision-based method for determining mouse body mass.
- To assess the feasibility of non-invasive, continuous mass monitoring in rodents.
- To reduce confounding factors associated with manual weighing in animal studies.
Main Methods:
- Utilized computer vision algorithms to analyze video data of mice.
- Combined video analysis with statistical modeling to predict body mass.
- Validated the method across diverse mouse strains, coat colors, and body masses.
Main Results:
- The computer vision method determined mouse body mass with a 4.8% error rate.
- The accuracy is sufficient to replace traditional manual weighing in most studies.
- Demonstrated feasibility across genetically diverse mouse populations.
Conclusions:
- Visually determining rodent mass via video analysis is a viable non-invasive technique.
- This method enables continuous monitoring, enhancing preclinical study quality.
- Improved animal welfare is a significant benefit of this non-invasive approach.
