Table 3.
Performance of different methods in MDD/HC classification with 10-fold cross-validation.
| Method | ACC(%) | SEN(%) | SPE(%) | f1(%) | AUC(%) |
|---|---|---|---|---|---|
| Nearest Neighbors | 54.5(0.9) | 55.9(0.9) | 53.1(0.9) | 56.1(0.9) | 54.5(0.9) |
| AdaBoost | 54.6(1.7) | 55.8(1.5) | 53.3(1.8) | 56.6(1.8) | 54.5(1.7) |
| Naive Bayes | 59.2(0.9) | 62.1(1.0) | 56.8(0.8) | 57.3(1.0) | 59.4(0.9) |
| Gaussian Process | 60.4(0.6) | 61.1(0.8) | 59.6(0.5) | 62.4(0.4) | 60.3(0.6) |
| Linear SVM | 62.8(0.7) | 61.9(0.6) | 64.2(0.9) | 66.7(0.7) | 62.5(0.7) |
| Deep Neural Net | 64.2(0.9) | 64.4(1.0) | 64.1(1.2) | 66.3(1.2) | 64.1(0.9) |
| GAN | 70.1(0.6) | 73.5(4.7) | 66.5(4.7) | 71.7(1.5) | 70.3(0.9) |
Table 3 The performance of different methods in MDD/HC classification with 10-fold cross-validation. SVM, support vector machine; GAN, Generative Adversarial Networks; ACC, Accuracy; SEN, sensitivity; SPE, specificity; F1, F-score; AUC, area under curve; MDD: major depressive disorder; HC: healthy control.