Table 3. Area Under the ROC curve results with 95% confidence intervals for the test set (n = 11,997).

Best AUCs for logistic regression and DNN are in bold.

Clinical Risk Score	AUC (95% CI)
Surgical Apgar	0.58 (0.52 – 0.64)
POSPOM	0.74 (0.68 – 0.79)
ASA	0.84 (0.80 – 0.87)
RQI Score*	0.91 (0.87 – 0.94)
RSI Uncalibrated**	0.97 (0.94 – 0.99)

	AUC (95% CI)
Model	With All 87 features	With ASA (88 features)	With POSPOM (88 features)	With Reduced Feature Set (45 features)	With Reduced Feature Set & ASA (46 features)	With Reduced Feature Set & POSPOM (46 features)
Logistic Regression	0.86 (0.81 – 0.90)	0.89 (0.86 – 0.92)	0.89 (0.85 – 0.93)	0.86 (0.81 – 0.90)	0.90 (0.87 – 0.93)	0.90 (0.86 – 0.93)
DNN	0.88 (0.85 – 0.91)	0.90 (0.87 – 0.93)	0.91 (0.87 – 0.95)	0.89 (0.85 – 0.92)	0.91 (0.88 – 0.93)	0.90 (0.87 – 0.93)

It should be noted that

^*RQI was calculated on 5,591 test patients (38 mortality) and

^**RSI Uncalibrated was calculated on 11,939 test patients (86 mortality)

AUC: Area under the curve; POSPOM: Preoperative Score to Predict Postoperative Mortality; RQI: Risk Quantification Index; RSI: Risk Stratification Index; DNN: Deep neural network.