TABLE 4.

Summary of performance metrics for left upper quadrant AI model.

LUQ Results	Abdominal Hemorrhage	Hemothorax	Negative
	Average ± StDev	Average ± StDev	Average ± StDev
Precision	0.980 ± 0.029	0.951 ± 0.049	0.982 ± 0.015
Recall	0.946 ± 0.058	0.992 ± 0.011	0.971 ± 0.025
F1	0.962 ± 0.033	0.971 ± 0.026	0.976 ± 0.018
Accuracy	0.975 ± 0.021	0.980 ± 0.019	0.984 ± 0.012
Specificity	0.990 ± 0.015	0.973 ± 0.028	0.990 ± 0.015
AUC a	0.997 ± 0.003	0.997 ± 0.002	0.998 ± 0.002

Average results and standard deviations are shown for n = 3 trained models for each classification category: abdominal hemorrhage positive, hemothorax positive, and negative for both injuries.

^a Area Under the ROC (receiver operating characteristic) Curve.