Automatic Cough Counting System
Author Information
Author(s): Barry Samantha J, Dane Adrie D, Morice Alyn H, Walmsley Anthony D
Primary Institution: University of Hull
Hypothesis
Can an automated system accurately recognize and count coughs from audio recordings?
Conclusion
An automated system for analyzing cough sounds has been developed, demonstrating high accuracy and efficiency.
Supporting Evidence
- HACC reduced cough counting time by 97.5%.
- The system achieved a sensitivity of 80% and specificity of 96%.
- Reproducibility of HACC analysis was 100%.
Takeaway
Researchers created a computer program that can listen to cough sounds and count them much faster than a person can.
Methodology
The Hull Automatic Cough Counter (HACC) uses digital signal processing and a probabilistic neural network to classify cough and non-cough sounds from audio recordings.
Potential Biases
The program may misclassify non-subject coughs as subject coughs due to background noise.
Limitations
The system cannot distinguish between coughs from the subject and ambient coughs.
Participant Demographics
33 smoking subjects, 20 male and 13 female, aged 20 to 54 with chronic cough.
Statistical Information
P-Value
p ≤ 0.05
Statistical Significance
p ≤ 0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website