Performance of the decision stump classifier | ||||||
Evaluation method | Sensitivity | Accuracy | Specificity | AUC | PPVa | NPVa |
10-fold cross validation | 67.2Â % | 73.4Â % | 74.9Â % | 0.710 | 38.1Â % | 91.0Â % |
testing on each patient’s last assessment | 51.1 % | 73.9 % | 82.0 % | 0.665 | 50.0 % | 82.6 % |
Performance of the six advanced classifiers measured by the 10-fold cross validation method | ||||||
Classifier | Sensitivity | Accuracy | Specificity | AUC | PPVa | NPVa |
Multiboost with decision stumps | 73.8Â % | 71.8Â % | 71.4Â % | 0.761 | 37.1Â % | 92.4Â % |
Support vector machine | 71.5Â % | 72.0Â % | 72.0Â % | 0.718 | 37.0Â % | 91.8Â % |
Deep learning | 71.6Â % | 72.3Â % | 72.5Â % | 0.744 | 37.2Â % | 91.8Â % |
Naive Bayes | 59.8Â % | 78.1Â % | 82.3Â % | 0.777 | 43.7Â % | 90.0Â % |
k-nearest neighbor | 56.9Â % | 73.3Â % | 77.0Â % | 0.704 | 36.0Â % | 88.7Â % |
Random forest | 48.3Â % | 75.8Â % | 82.0Â % | 0.662 | 37.9Â % | 87.5Â % |
Performance of the six advanced classifiers measured by the method of testing on each patient’s last assessment | ||||||
Classifier | Sensitivity | Accuracy | Specificity | AUC | PPVa | NPVa |
Multiboost with decision stumps | 74.5Â % | 74.4Â % | 74.4Â % | 0.757 | 50.7Â % | 89.2Â % |
Support vector machine | 70.2Â % | 73.3Â % | 74.4Â % | 0.723 | 49.3Â % | 87.6Â % |
Deep learning | 68.1Â % | 72.2Â % | 73.7Â % | 0.738 | 47.8Â % | 86.7Â % |
Naive Bayes | 44.7Â % | 73.9Â % | 84.2Â % | 0.783 | 50.0Â % | 81.2Â % |
k-nearest neighbor | 48.9Â % | 73.9Â % | 82.7Â % | 0.773 | 50.0Â % | 82.1Â % |
Random forest | 38.3Â % | 75.6Â % | 88.7Â % | 0.678 | 54.5Â % | 80.3Â % |