Skip to main content
. 2021 Apr 29;12:670944. doi: 10.3389/fneur.2021.670944

Table 3.

Comparison of the classification of the classification and regression trees algorithm with the diagnosis made at the German Center for Vertigo and Balance Disorders (DSGZ) for functional dizziness (FD), vestibular migraine (VM), Menière's disease (MD), benign paroxysmal positional vertigo (BPPV), unilateral vestibulopathy (UVP), bilateral vestibulopathy (BVP), and vestibular paroxysmia (VP).

Diagnosis by DSGZ Diagnostic parameters
FD VM MD BPPV UVP BVP VP SENS (%) SPEC (%) PPV (%) NPV (%)
Diagnosis by classification algorithm FD 146 41 5 8 14 7 2 42.2 57.8 65.5 34.5
VM 63 84 22 22 16 6 6 39.1 60.9 38.4 61.6
MD 30 28 70 7 15 9 0 49.3 50.7 44.0 56.0
BPPV 36 17 20 76 13 11 12 56.7 43.3 41.1 58.9
UVP 8 11 4 12 23 0 1 20.2 79.8 39.0 61.0
BVP 43 17 19 6 28 32 9 48.5 51.5 20.8 79.2
VP 20 17 2 3 5 1 19 38.8 61.2 28.4 71.6

Performance is described by SENS, sensitivity; SPEC, specificity; PPV, positive predictive value, and NPV, negative predictive value.