Skip to main content

Table 1 The different sets of imaging feature combinations used as input to the approach

From: A three-stage, deep learning, ensemble approach for prognosis in patients with Parkinson’s disease

  Feature set combinations
1 DaTscan + Semi-quantitative + All ImageNet
2 DaTscan + Semi-quantitative
3 DaTscan + All ImageNet
4 Semi-quantitative + All ImageNet
5 DaTscan
6 Semi-quantitative
7 All ImageNet
8 VGG16
9 ResNet50
10 DenseNet121
11 InceptionV3
  1. DaTscan imaging features refer to the imaging features extracted directly from the convolutional LSTM network. Semi-quantitative refers to the striatal binding ratio values from the right and left caudate nuclei and putamina