Diagnosing Coronary Heart Disease with Qi Deficiency Syndrome Using Data Mining
Author Information
Author(s): Zhao Huihui, Chen Jianxin, Hou Na, Zhang Peng, Wang Yong, Han Jing, Hou Qin, Qi Qige, Wang Wei
Primary Institution: Beijing University of Chinese Medicine
Hypothesis
Can a data mining method accurately establish a diagnosis pattern for coronary heart disease with Qi deficiency syndrome?
Conclusion
The study established a diagnosis pattern for coronary heart disease with Qi deficiency syndrome using five biological parameters, achieving a diagnosis accuracy of 84.5%.
Supporting Evidence
- The diagnosis accuracy of the patterns could reach 84.5% based on a 3-fold cross validation technique.
- The classification accuracy is 83.5% when tested on an additional 85 unstable angina cases.
- The association between symptoms and the five biological parameters reached an accuracy of ∼80%.
Takeaway
Doctors can use a special computer method to help figure out if someone has a heart problem related to a lack of energy, using just a few blood tests.
Methodology
The study used a t-test-based Adaboost algorithm to analyze 34 biological parameters in 52 patients with unstable angina.
Potential Biases
Potential bias from the subjective diagnosis criteria used by TCM doctors.
Limitations
The study did not include a health control group and focused only on unstable angina patients.
Participant Demographics
Patients aged between 55 and 75 years, with 39 having Qi deficiency syndrome and 13 without.
Statistical Information
P-Value
0.023899 for MCH, 0.027355 for CHO
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website