New Algorithm for Classifying Floating and Sunken Pulses
Author Information
Author(s): Kim Jaeuk U., Jeon Young Ju, Lee Yu Jung, Kim Keun Ho, Kim Jong Yeol
Primary Institution: Korea Institute of Oriental Medicine
Hypothesis
Can a novel classification algorithm accurately differentiate between floating and sunken pulse qualities in pulse diagnosis?
Conclusion
The study confirms the validity of a new classification algorithm for floating and sunken pulses, achieving a concordance rate of 72% between diagnoses made by the algorithm and oriental medical doctors.
Supporting Evidence
- 121 out of 169 diagnoses were concordant between the algorithm and the doctors.
- The accuracy of the algorithm reached 72% with a Matthews correlation coefficient of 0.42.
- Subjects with sunken pulses had significantly higher BMI compared to those with floating pulses.
- The algorithm's performance was validated through clinical tests involving multiple doctors.
Takeaway
Doctors can use a new method to tell if a pulse is floating or sunken, which helps in understanding a person's health better.
Methodology
The study involved 169 healthy volunteers diagnosed by 12 oriental medical doctors using a new algorithm based on pulse strength measurements at different pressures.
Potential Biases
Potential bias from subjective interpretations by the oriental medical doctors despite efforts to reduce it through paired diagnoses.
Limitations
The study was limited to healthy volunteers in their 20s, which may not represent the general population.
Participant Demographics
169 healthy volunteers, aged 20s, with no vascular deformity.
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