Automatic Detection of Aspiration in Children with Dysphagia
Author Information
Author(s): Lee Joon, Blain Stefanie, Casas Mike, Kenny Dave, Berall Glenn, Chau Tom
Primary Institution: Bloorview Kids Rehab, Toronto, Ontario, Canada
Hypothesis
Can a classifier accurately detect aspiration in children with dysphagia using non-invasive methods?
Conclusion
The proposed aspiration classification algorithm provides promising accuracy for aspiration detection in children.
Supporting Evidence
- The classifier achieved an adjusted accuracy of 81.3 ± 8.5% for aspiration detection.
- The feature pairing of dispersion ratio and normality was particularly effective.
- The classifier's performance was statistically comparable to existing methods.
Takeaway
Researchers created a device that can tell if kids are swallowing safely or if they might be choking, helping caregivers keep children safe during meals.
Methodology
Vibration signals from children were recorded and analyzed using a radial basis function classifier to distinguish between safe swallows and aspirations.
Potential Biases
The sample size may limit the generalizability of the findings.
Limitations
The classifier was based solely on vibration signals and did not incorporate other physiological indicators.
Participant Demographics
Mean age of participants was 6.0 years, with 64 males and 53 females, all with neurological dysphagia.
Statistical Information
P-Value
p ≤ 8.5 × 10-4
Confidence Interval
79.8 ± 7.3%
Statistical Significance
p > 0.2
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website