Automated and Continuous Monitoring of Animal Welfare through Digital Alerting

Area of Science:

• Preclinical research
• Animal welfare science
• Biomedical research

Background:

• Preclinical animal research prioritizes welfare and data integrity.
• Current welfare monitoring relies on manual clinical observations.
• Increased handling for assessments can cause animal stress and data variability.

Purpose of the Study:

• To investigate automated alerting for animal welfare.
• To determine if behavior and physiology monitoring can improve distress identification.
• To complement existing manual welfare assessments.

Main Methods:

• Developed algorithms using historical data to detect motion and breathing rate changes.
• Analyzed data retrospectively over a 31-day period.
• Evaluated algorithm performance on animal models of rapid aging and lung metastasis.

Main Results:

• Automated alerts frequently preceded or coincided with technician observations of declining health.
• Algorithms demonstrated sensitivity and accuracy in identifying unhealthy animals.
• Alert percentages varied by animal model (0-24% of subject days).

Conclusions:

• Automated alerting effectively complements standard clinical observations.
• This technology enhances animal welfare and promotes responsible scientific practices.
• Automated systems aid in timely identification of animals needing humane endpoints.