The signal decomposition feature dataset of WD patients and healthy individuals
收藏资源简介:
The three files features_DMFCC.pkl, features_HCC.pkl, and features_MFCC.pkl are the improved Mel-frequency Cepstrum Coefficient using signal decomposition feature file, Hilbert Cepstrum Coefficient feature file, and Mel-frequency Cepstrum Coefficient feature file, respectively. The feature data file contains 120 native Mandarin-speaking participants, including 60 WD patients and 60 sex-matched healthy individuals. The first five columns and the last column of data for all the feature data files are ID number, label (0 for healthy individuals and 1 for WD patients), age, sex one-hot code (gender_0=1 for females and gender_1=1 for males), and Frenchay dysarthria scale scores, respectively. The features_DMFCC.pkl and features_HCC.pkl files have VMD, CEEMDAN, and EMD feature columns representing complete ensemble EMD with adaptive noise features, variational mode decomposition features, and empirical mode decomposition features, respectively. The features_MFCC.pkl file has MFCC feature columns representing Mel-frequency cepstrum coefficients features.



