Development and internal validation of machine learning algorithms to predict patient satisfaction after total hip arthroplasty

Table 3 Model performance for predicting patient satisfaction on test set (n = 302)

	LASSO	SVM	RF	XGB
AUC	0.76 (0.67–0.86)	0.74 (0.63–0.85)	0.68 (0.56–0.80)	0.66 (0.50–0.78)
Brier score	0.23	0.23	0.23	0.21
Threshold	0.50	0.52	0.53	0.58
Sensitivity	65.3%	57.9%	53.0%	50.5%
Specificity	82.4%	76.5%	76.5%	76.5%
Calibration slope	1.29	0.44	1.06	0.29
Calibration intercept	0.13	0.70	0.40	0.77

ISSN: 2524-7948