Research Article
SchizoGoogLeNet: The GoogLeNet-Based Deep Feature Extraction Design for Automatic Detection of Schizophrenia
Table 3
Overall performances of the proposed methods for the balanced dataset.
| Classifier | Sensitivity (%) | Specificity (%) | Accuracy (%) | Positive predictive value (%) | F1-score (%) |
| GoogLeNet | 92.15 | 96.42 | 94.28 | 96.26 | 94.16 | SVM | 98.33 | 98.27 | 98.30 | 98.27 | 98.30 | KNN | 97.29 | 97.27 | 97.28 | 97.27 | 97.28 | DT | 97.51 | 97.21 | 97.36 | 97.22 | 97.36 | LDA | 98.53 | 98.07 | 98.29 | 98.08 | 98.30 |
|
|