Research Article
Human Tongue Thermography Could Be a Prognostic Tool for Prescreening the Type II Diabetes Mellitus
Table 6
Comparison of classification performance of SVM, NB, and CNN machine learning algorithms.
| | Classifiers | AUC | Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | Accuracy (%) |
| | Support vector machine (SVM) | 0.92 | 90 | 95.71 | 91.30 | 90.54 | 92.85 | | Naïve Bayes (NB) | 0.96 | 95.71 | 82.85 | 84.81 | 95.08 | 89.28 | | Convolutional Neural Network (CNN-VGG16 net) | 0.80 | 92.85 | 95.71 | 95.58 | 93.05 | 94.28 |
|
|
AUC: area under the curve (ROC), PPV: positive predictive value, NPV: negative predictive value.
|
