Skip to main content

Table 4 Diagnostic accuracy of NNets for predicting clinical end-points after attribute selection in the test cohort

From: Computational classifiers for predicting the short-term course of Multiple sclerosis

 

End-points

A% (SD)

S

%

Sp

%

NPV

%

AUC% (SD)

PPV

%

Kappa

 

Change EDSS

80 (14)

92

61

80

76 (25)

80

0.54

NNets

Disability progression

75 (17)

87

52

61

74 (31)

80

0.37

 

Relapse-free

67 (21)

53

77

70

65 (22)

61

0.33

  1. The results are expressed as the percentage (SD). The attributes selected for the different classifiers (see Table S4) were: 1) Clinical variables: disease subtype, age, sex, EDSS, motor score of the EDSS, MSFC, NHPT, TWT; 2) MRI variables: T1 lesion volume, Gad+ lesion volume, GM volume, WM volume; 3) MEP variables: presence of pathological MEP, MEP score, CMCT.
  2. SD: Standard Deviation; AUC: Area under ROC curve; A: Accuracy; S: sensitivity; Sp: specificity; PPV: positive predictive value; NPV: negative predictive value.
Back to article page

BMC Neurology

ISSN: 1471-2377

Contact us