Table 2 Performance of the gauze EBL regression models on the porcine test set.
From: Automated deep learning model for estimating intraoperative blood loss using gauze images
Base model | CS model | P-W CS model | ||||
|---|---|---|---|---|---|---|
MAE (g) | MAPE (%) | MAE (g) | MAPE (%) | MAE (g) | MAPE (%) | |
Performance of the gauze EBL regression models evaluated on the test set (n = 62) | ||||||
EBL value (x) | ||||||
\({\text{x}}\le 3{\text{g}}\) | 0.31 | 10.36 | 0.28 | 9.38 | 0.21 | 6.96 |
\(3{\text{g}}<{\text{x}}\le 5{\text{g}}\) | 0.38 | 7.06 | 0.32 | 5.98 | 0.23 | 4.34 |
\(\mathrm{ x}>5{\text{g}}\) | 0.71 | 9.10 | 0.46 | 6.15 | 0.43 | 5.72 |
Average | 0.46 | 8.56 | 0.35 | 6.97 | 0.28 | 5.46 |
