Skip to main content
. 2023 Feb 6;12(4):1297. doi: 10.3390/jcm12041297

Table 5.

ICC of different models and pressure types.

95% Confidence Interval F Test with True Value 0
Task and Method ICC b Lower Bound Upper Bound Value Sig.
mPAP and CatBoost
Single Measures 0.525 a −0.289 0.898 3.211 0.091
Average Measures 0.689 c −0.813 0.946 3.211 0.091
mPAP and SVM
Single Measures 0.074 a −0.668 0.742 1.160 0.431
Average Measures 0.138 c −4.015 0.852 1.160 0.431
mPAP and XGBoost
Single Measures 0.877 a 0.447 0.978 15.211 0.002
Average Measures 0.934 c 0.617 0.989 15.211 0.002
dPAP and XGBoost
Single Measures 0.824 a 0.280 0.967 10.355 0.006
Average Measures 0.903 c 0.438 0.983 10.355 0.006
sPAP and XGBoost
Single Measures 0.963 a 0.803 0.994 53.139 0.000
Average Measures 0.981 c 0.890 0.997 53.139 0.000
TPR and XGBoost
Single Measures 0.521 a −0.294 0.897 3.173 0.093
Average Measures 0.685 c −0.834 0.946 3.173 0.093

Note. Two-way mixed effects model where people effects are random and measures effects are fixed; a The estimator is the same, whether the interaction effect is present or not; b Type C intraclass correlation coefficient using a consistency definition. The between-measure variance is excluded from the denominator variance; c This estimate is computed assuming the interaction effect is absent, because it is not estimable otherwise.