Table 1 Machine learning to evaluate the diagnostic significance of ROI activity.
From: Hyperactive sensorimotor cortex during voice perception in spasmodic dysphonia
| Â | Accuracy (%) | Precision (%) | Recall (%) | Specificity (%) | F-measure (%) |
|---|---|---|---|---|---|
Linear discriminant analysis | |||||
Sensorimotor | 73 | 69 | 82 | 64 | 75 |
SMA | 59 | 58 | 64 | 55 | 61 |
Thalamus | 27 | 27 | 27 | 27 | 27 |
Cerebellum | 41 | 42 | 45 | 36 | 43 |
Putamen | 32 | 30 | 27 | 36 | 29 |
Pallidum | 23 | 25 | 27 | 18 | 26 |
Support vector machine | |||||
Sensorimotor | 59 | 60 | 55 | 64 | 57 |
SMA | 50 | 50 | 55 | 45 | 52 |
Thalamus | 59 | 60 | 55 | 64 | 57 |
Cerebellum | 50 | 50 | 55 | 45 | 52 |
Putamen | 30 | 31 | 36 | 25 | 33 |
Pallidum | 32 | 33 | 36 | 27 | 35 |
Logistic regression | |||||
Sensorimotor | 68 | 67 | 73 | 64 | 70 |
SMA | 57 | 54 | 64 | 50 | 58 |
Thalamus | 32 | 33 | 36 | 27 | 35 |
Cerebellum | 43 | 42 | 45 | 42 | 43 |
Putamen | 32 | 30 | 27 | 36 | 29 |
Pallidum | 23 | 25 | 27 | 18 | 26 |