Related Experiment Video
Updated: Sep 14, 2025
04:04
Asthma Detection Research Based on Voice Signal Processing and Machine Learning
Published on: July 22, 2025
420
Artificial intelligence in smartphone video analysis for equine asthma diagnostic support
Carolina Gomes1,2,3, Luísa Coheur1,2, Paula Tilley3,4
1Instituto Superior Técnico, Universidade de Lisboa, Lisbon, Portugal.
Equine Veterinary Journal
|July 21, 2025
Summary
A new AI tool, EquiBreathe, shows promise for non-invasively screening equine asthma. This cost-effective solution aids veterinarians and handlers in early detection, improving horse health and performance.
Area of Science:
- Veterinary Medicine
- Animal Health
- Artificial Intelligence in Diagnostics
Background:
- Equine asthma significantly impacts horse health and performance.
- Current diagnostic methods are invasive and require specialized equipment.
- A non-invasive, cost-effective screening tool is needed for ambulatory settings.
Purpose of the Study:
- Assess willingness to adopt a new diagnostic tool.
- Identify challenges in visual equine asthma recognition.
- Develop EquiBreathe, an AI-powered non-invasive diagnostic tool.
Main Methods:
- Cross-sectional survey using Google Forms questionnaires.
- AI model development using feature engineering and image subtraction.
- Video analysis of nostril and abdominal movements in 23 horses.
Main Results:
- High interest in the AI tool from veterinarians and handlers.
- Visual assessment alone is challenging for equine asthma diagnosis.
- The AI model achieved 89% accuracy in detecting equine asthma.
Conclusions:
- A field-friendly diagnostic solution for equine asthma is needed.
- EquiBreathe shows potential as a non-invasive screening tool.
- The AI tool can improve diagnostic precision and communication.
