Smart Garment for Continuous Respiration Monitoring in Canines.

Area of Science:

• Veterinary Medicine
• Biomedical Engineering
• Wearable Technology

Background:

• Canines require continuous respiration monitoring for health insights, but current methods are invasive or uncomfortable.
• Breed-specific anatomy and active lifestyles contribute to canine respiratory issues.

Purpose of the Study:

• To develop a smart garment for continuous, non-invasive monitoring of canine respiration, body temperature, and physical activity.
• To validate the system's accuracy across different breeds and activity levels.

Main Methods:

• Integration of a spongy strain sensor and data acquisition module into canine apparel.
• Validation using Labrador and Boxer breeds to capture respiration patterns.
• Application of convolutional neural network (CNN)-assisted machine learning (ML) for pattern classification.

Main Results:

• The smart garment successfully captured breed- and activity-specific respiration patterns, including breathing rate, amplitude, and panting.
• The CNN-ML system achieved over 94.3% accuracy in classifying respiratory patterns.
• Demonstrated potential for monitoring body temperature and physical activity alongside respiration.

Conclusions:

• The smart garment provides a viable solution for continuous, non-invasive at-home canine health monitoring.
• This platform shows promise for broader applications in companion animal veterinary care, including cats.