Prediction of Parkson disease by speech recognition based on pytorch
Algorithm design based on GAN principle Parkinson2
You can get the data
|
prediction |
F1 |
MCC |
Concat |
69.04 |
64.99 |
0.3379 |
TQWT |
80.78 |
76.24 |
0.5683 |
MFCC |
82.5 |
88.6 |
0.513 |
Wavelet |
75 |
85.1 |
0.161 |
|
prediction |
recall |
F1 |
MCC |
Concat_Wavelet |
67.92 |
65.49 |
66.68 |
0.3333 |
MFCC_Concat |
69.19 |
67.06 |
68.11 |
0.3618 |
MFCC_Wavelet |
64.47 |
61.58 |
62.99 |
0.2589 |
TQWT_Concat |
79.56 |
75.18 |
77.31 |
0.5457 |
TQWT_MFCC |
88.36 |
81.22 |
84.64 |
0.6921 |
TQWT_Wavelet |
77.8 |
75.68 |
76.73 |
0.5344 |
|
prediction |
recall |
F1 |
MCC |
TQWT_MFCC_Wavelet |
75.87 |
72.04 |
73.91 |
0.4776 |
TQWT_MFCC_Concat |
84.9 |
78.85 |
81.76 |
0.6347 |
TQWT_Wavelet_Concat |
80.07 |
77.78 |
78.91 |
0.5780 |
MFCC_Wavelet_Concat |
68.44 |
65.24 |
66.8 |
0.3353 |