Table 3. Prediction table for each phenotype and associated classification error, as obtained by employing evtree on TD.
Each line in the table shows (in bold characters) how many patients in TD are correctly classified in the corresponding phenotype.
| Patients actually belonging to phenotype | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Ph 1 | Ph 2 | Ph 3 | Ph 4 | Ph 5 | Ph 6 | Ph 7 | Ph 8 | Error | ||
| Patients predicted in | Ph 1 | 294 | 0 | 0 | 103 | 60 | 0 | 0 | 0 | 35.67% |
| Ph 2 | 0 | 116 | 0 | 42 | 0 | 0 | 0 | 0 | 26.58% | |
| Ph 3 | 0 | 0 | 163 | 11 | 12 | 1 | 0 | 0 | 12.36% | |
| Ph 4 | 0 | 0 | 8 | 17 | 1 | 0 | 0 | 0 | 34.61% | |
| Ph 5 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 100% | |
| Ph 6 | 0 | 0 | 0 | 33 | 3 | 61 | 0 | 0 | 37.11% | |
| Ph 7 | 0 | 0 | 7 | 0 | 0 | 0 | 216 | 0 | 3.14% | |
| Ph 8 | 0 | 0 | 0 | 2 | 0 | 0 | 3 | 218 | 2.24% | |