A Friedman test revealed significant overall differences between models across voxel- and surface-based metrics, including Dice, IoU, Accuracy, Precision, HD, and ASD (all p < 0.01), while Recall did not reach statistical significance (p = 0.057).
← all excerpts
Synthetically trained convolutional neural networks for time-resolved aortic segmentation of four-dimensional flow magnetic resonance imaging.
1
0.0570
0.0570