Table 1 Cross validation performance for the macaque data (N = 58 from 22 subjects)
From: Deep learning-based analysis of macaque corneal sub-basal nerve fibers in confocal microscopy images
Architecture | 5-Fold CV R2 | Number of trainable parameters |
|---|---|---|
Autoencoder | 0.733 | 1,330,498 |
U-Net (deepNerve) | 0.859 | 487,730 |
U-Net with Residual Branches | 0.731 | 516,578 |
ACCMetrics | 0.718 | N/A |