Table 2 Comparison of area under the curve for identification of the presence of microcirculatory alterations associated with COVID-19 status between the algorithm-based, deep learning-based and combined models in internal and external validation
Model type | AUROC (CI) for identification of COVID-19 status | Linear regression estimates as mean difference ± S.E. (CI) | T statistic and P value | |||
|---|---|---|---|---|---|---|
Internal validation cohort | External validation cohort | Between-model comparison | Between-cohort comparison | Between-model comparison | Between-cohort comparison | |
Algorithm-based model | 0.74 (0.69–0.79) | 0.73 (0.71–0.76) | – |  − 0.10 ± 0 .00 (− 0.10 to − 0.10) | – |  − 61.88, < 0.0001 |
Deep learning-based model | 0.81 (0.76–0.86) | 0.61 (0.58–0.63) |  − 0.03 ± 0.00 (− 0.03 to − 0.03) |  − 15.04, < 0.0001 | ||
Combined model | 0.84 (0.80–0.89) | 0.75 (0.73–0.78) | 0.06 ± 0.00 (0.06–0.06) | 30.20, < 0.0001 | ||