Skip to main content

Table 4 Performance on test sets using gait metrics from the electronic walkway (in %)

From: Detecting cognitive impairment in cerebrovascular disease using gait, dual tasks, and machine learning

Condition

SS

DAn

DS1

DS7

F

DAn + SSa)

DS1 + SSa)

DS7 + SSb)

F + SSb)

Classifier

LR

SVM

LR

SVM

RF

SVM

SVM

SVM

LR

Feature Sel.

RFE

RFE

RFE

None

RF

None

KBest

RF

RFE

Sampling

SMOTE

SMOTE

None

SMOTE

None

None

None

SMOTE

SMOTE

# Features

2.0 ± 2.2

3.2 ± 2.2

2.2 ± 1.8

12.0 ± 0.0

6.0 ± 0.8

12.0 ± 0.0

1.7 ± 2.1

8.6 ± 1.3

2.8 ± 2.1

Bal. Acc.

76.8

75.1

71.6

77.5

60.5

74.0

78.8

75.1

69.5

Sensitivity

75.9

72.4

65.5

82.8

65.5

75.9

96.6

72.4

72.4

Specificity

77.8

77.8

77.8

72.2

55.6

72.2

61.1

77.8

66.7

F1 score

80.0

77.8

73.1

82.8

67.9

78.6

87.5

77.8

75.0

AUC

75.3

75.9

69.2

79.5

54.8

73.9

68.6

72.4

67.2

  1. SS: Single-task; DAn: Walking while naming animals; DS1: Serial subtraction by 1s; DS7: Serial subtraction by 7s; F: Walking Fast; LR: Logistic Regression; RF: Random Forest; SVM: Support Vector Machine; RFE: Recursive feature elimination; AUC: Area under the curve. a) using dual-task costs; b) using dual-task costs (or capacity indexes) and SS metrics