Research Article
A Urine Metabonomics Study of Rat Bladder Cancer by Combining Gas Chromatography-Mass Spectrometry with Random Forest Algorithm
Table 2
Classification group for three models.
| Results | Sensitivity (%) | Specificity (%) | Accuracy (%) | MCC | AUC |
| Healthy control vs. early- and middle-stage group | 83.33 | 98.80 | 96.06 | 0.8609 | 0.9775 | Healthy control vs. advanced-stage group | 97.22 | 97.60 | 97.54 | 0.9194 | 0.9868 | Early- and middle-stage groups vs. advanced stage group | 80.56 | 97.01 | 94.09 | 0.7909 | 0.9546 |
|
|
MCC: Matthews correlation coefficient, AUC: area under curve.
|