Table 4 Comparative analysis of models' performance across all folds.
From: Deep learning approaches for assessing pediatric sleep apnea severity through SpO2 signals
Model | Average four-class accuracy (%) | Average four-class kappa | Model size (MB) | Number of parameters |
|---|---|---|---|---|
ResNet | 72.90 | 0.57 | 85.32 | 22,366,673 |
CNN-BiGRU-Attention | 75.95 | 0.63 | 0.199 | 51,089 |
