Table 3 Classification performance of TGNet on unimodal and multimodal brain network datasets
From: TGNet: tensor-based graph convolutional networks for multimodal brain network analysis
Method | Dataset | Modality | Accuracy | AUC |
|---|---|---|---|---|
TGNet | HIV | fMRI | 53.61±11.88 | 52.31±13.63 |
DTI | 75.71±15.12 | 76.08±15.61 | ||
Both | 81.39±13.41 | 82.08±14.81 | ||
BP | fMRI | 57.78±12.61 | 51.83±10.84 | |
DTI | 64.44±13.59 | 63.34±15.91 | ||
Both | 67.78±12.28 | 66.31±10.24 | ||
PPMI | PICo | 79.01±6.62 | 73.66±8.69 | |
Hough | 80.14±5.65 | 72.03±7.42 | ||
Probtracx | 83.39±4.34 | 76.50±8.29 | ||
All | 83.94±6.09 | 77.93±6.86 |