Table 4 The performances of SVM classifiers based on different kinds of features
From: A study of hepatic fibrosis staging methods using diffraction enhanced imaging
Feature Set | Accuracy | Sensitivity | Specificity | Training time (s) | Prediction speed (obj/s) |
|---|---|---|---|---|---|
FO | 98.41% | 98.68% | 97.39% | 58.413 | 12,000 |
GLCM | 99.49% | 99.73% | 98.56% | 93.538 | 6900 |
GGCM | 99.31% | 99.35% | 99.15% | 61.643 | 11,000 |
FO_GLCM | 99.73% | 98.85% | 99.25% | 40.354 | 13,000 |
FO_ GGCM | 99.68% | 99.75% | 99.40% | 45.398 | 13,000 |
GLCM_GGCM | 99.73% | 99.82% | 99.37% | 41.978 | 12,000 |
FO_GLCM_GGCM | 99.81% | 99.84% | 99.69% | 48.903 | 10,000 |