Table 4 Comparative analysis of pre-trained models with frozen convolutional layers.

From: Design of a deep fusion model for early Parkinson’s disease prediction using handwritten image analysis

Model

Precision

Recall

F1-measure

MCC

AlexNet

92.46

92.42

92.39

91.92

DenseNet 201

96.85

96.80

96.82

96.60

SqueezeNet1.1

90.32

90.34

90.22

89.87

VGG-16 Net

90.08

89.82

89.77

89.17

VGG-19 Net

90.85

90.80

90.82

90.60

ResNet-50

95.09

94.91

94.93

94.58

ResNet-101

95.75

95.39

95.43

95.10

MobileNetV1

96.56

96.24

96.45

96.34

MobileNetV2

97.56

97.34

97.45

97.10

GoogleNet

94.68

94.01

94.53

94

Back to article page