Table 1 Table of results

From: Exploiting macro- and micro-structural brain changes for improved Parkinson’s disease classification from MRI data

Model

ROC-AUC

ACC

SPE

SEN

CNNCOMBINED

0.89

80.8%

82.4%

79.1%

CNNDTI

0.86

78.1%

79.3%

76.9%

CNNFA

0.86

81.3%

78.3%

82.6%

CNNMD

0.85

77.1%

79.3%

74.7%

CNNRD

0.84

77.6%

77.3%

78.0%

CNNAD

0.85

77.6%

79.3%

75.8%

CNNJACa(P 0.032)

0.83

73.9%

67.0%

81.3%

CNNAFFa(P 0.035)

0.85

76.0%

85.5%

65.9%

RFCOMBINED (159 FEATURES)

0.85

79.7%

76.2%

79.8%

  1. ROC-AUC area under the receiver operating characteristic curve, ACC accuracy, SPE specificity, SEN sensitivity.
  2. aSignificant difference (P < 0.05) with CNNCOMBINED.