Table 7. Accuracy results of repeated 10 fold cross validation of the Random Forest and Linear Discriminant Analysis (LDA) models built to classify patients with prostate cancer and cancer-free controls based on blood PSA levels and urinary VOCs.
Repeated 10 fold cross validation (%) | ||||||
Model | Min. | 1st.Qu. | Median | Mean | 3rd.Qu. | Max. |
Random Forest | 20.00 | 66.67 | 72.73 | 73.69 | 80.00 | 100.00 |
LDA | 22.22 | 58.89 | 63.64 | 64.85 | 72.73 | 100.00 |
Monte-Carlo 10 fold cross validation (%) | ||||||
Model | Min. | 1st.Qu. | Median | Mean | 3rd.Qu. | Max. |
Random Forest | 10.00 | 45.45 | 55.56 | 55.79 | 66.67 | 90.91 |
LDA | 10.00 | 40.00 | 50.00 | 48.00 | 60.00 | 88.89 |