From: Immediate word recall in cognitive assessment can predict dementia using machine learning techniques
Cross-validation accuracy (%) | Accuracy (on data split) (%) | Precision | f1-score | Sensitivity | Specificity | ||
---|---|---|---|---|---|---|---|
Dementia | Normal | ||||||
A. Scenario 3A: on the sample person’s responses | |||||||
KNN | 93.00 | 93.00 | 0.77 | 0.94 | 0.92 | 0.36 | 0.99 |
Decision tree | 93.00 | 93.00 | 0.77 | 0.94 | 0.92 | 0.36 | 0.99 |
Random forest | 93.00 | 93.00 | 0.77 | 0.94 | 0.92 | 0.36 | 0.99 |
Artificial neural networks | 92.00 | 93.00 | 0.73 | 0.94 | 0.92 | 0.37 | 0.99 |
B. Scenario 3B: on the proxy’s responses | |||||||
KNN | 88.00 | 91.00 | 0.79 | 0.98 | 0.91 | 0.97 | 0.88 |
Decision tree | 88.00 | 92.00 | 0.80 | 0.99 | 0.92 | 0.98 | 0.89 |
Random forest | 88.00 | 92.00 | 0.80 | 0.99 | 0.92 | 0.98 | 0.89 |
Artificial neural networks | 88.00 | 91.00 | 0.83 | 0.95 | 0.91 | 0.89 | 0.92 |
C. Scenario 3C: on the sample person’s and proxy’s responses | |||||||
KNN | 95.00 | 94.00 | 0.89 | 1.00 | 0.94 | 1.00 | 0.88 |
Decision tree | 95.00 | 94.00 | 0.89 | 1.00 | 0.94 | 1.00 | 0.88 |
Random forest | 95.00 | 94.00 | 0.89 | 1.00 | 0.94 | 1.00 | 0.88 |
Artificial neural networks | 93.00 | 94.00 | 0.89 | 1.00 | 0.94 | 1.00 | 0.88 |