Highly Accurate and Precise Determination of Mouse Mass Using Computer Vision.

Area of Science:

• Biomedical research
• Animal science
• Computer vision

Background:

• Body mass changes are crucial health indicators in model organisms.
• Current manual weighing of rodents causes stress and misses rapid mass changes.
• A noninvasive, continuous method for monitoring animal mass is needed.

Approach:

• Utilized computer vision and statistical modeling to analyze mouse video data.
• Developed an image-based method to determine rodent body mass.
• Validated the approach across diverse mouse strains, coat colors, and masses.

Key Points:

• The image-based method achieved a 4.8% error rate in determining mouse mass.
• This accuracy is sufficient to replace manual weighing in many preclinical studies.
• The technique allows for noninvasive and continuous body mass assessment.

Conclusions:

• Visual determination of rodent mass using video is feasible.
• This noninvasive method enhances animal welfare and preclinical study reliability.
• Enables continuous monitoring, capturing rapid mass fluctuations vital for research.