TCM pulse analysis
Pulse diagnosis is widely used in Traditional Chinese Medicine (TCM) diagnosis. It is an effective method used by the TCM practitioner to understand the patient’s health condition. However, the accuracy of diagnosis result is subjective as it heavily dependent on the skills and experience of TCM pra...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/78837 |
| Summary: | Pulse diagnosis is widely used in Traditional Chinese Medicine (TCM) diagnosis. It is an effective method used by the TCM practitioner to understand the patient’s health condition. However, the accuracy of diagnosis result is subjective as it heavily dependent on the skills and experience of TCM practitioner. Research on quantitative methods in pulse diagnosis to better aid the analysis of the TCM pulses has been done using wavelet transform and curve fitting method followed by feature extraction and classification. Further assessment is required to ensure its effectiveness and reliability of the methods on more pulse samples.
This project aims to characterize the Fine pulse (细脉) and the Wiry pulse (弦脉) using curve fitting method to generate features parameters for Machine Learning classification. Two curve fitting models - Gamma Density Function and Gaussian Two-term Function was investigated to generate different set of feature parameter values for classification. Gamma function has a much better fitting result. Hence, it can be concluded that the Gamma Density Function has higher capability to fit and classify the pulse data set used by this project. Fisher’s ratio is used in the feature selection to select the features parameters for Classification. Features parameters generated by the Gamma Density function is selected for classification. The classification accuracy is more than 80% by using linear SVM training model to classify the Fine class and the Wiry class. |
|---|