Table 1b.

Summary of model metrics for each of the four machine learning techniques.

	Metrics	Minimum	5th Percentile	25th Percentile	Median	75th Percentile	95th Percentile	Maximum	Mean	SD	Range
XGBoost	Accuracy	0.684	0.751	0.773	0.789	0.805	0.828	0.898	0.789	0.024	0.215
	F1	0.686	0.748	0.771	0.787	0.803	0.826	0.897	0.787	0.024	0.210
	Sensitivity	0.680	0.764	0.787	0.802	0.818	0.840	0.901	0.802	0.023	0.222
	Specificity	0.592	0.750	0.773	0.789	0.805	0.827	0.947	0.789	0.024	0.354
	PPV	0.680	0.782	0.803	0.818	0.833	0.855	0.958	0.818	0.022	0.277
	NPV	0.567	0.720	0.744	0.761	0.777	0.801	0.930	0.761	0.025	0.363
	AUROC	0.772	0.836	0.855	0.868	0.881	0.900	0.948	0.868	0.020	0.176
Random Forest	Accuracy	0.675	0.744	0.768	0.784	0.800	0.823	0.892	0.784	0.024	0.216
	F1	0.687	0.746	0.769	0.785	0.801	0.824	0.884	0.785	0.024	0.198
	Sensitivity	0.665	0.755	0.777	0.793	0.809	0.832	0.895	0.793	0.023	0.229
	Specificity	0.584	0.731	0.754	0.771	0.787	0.811	0.927	0.771	0.024	0.344
	PPV	0.676	0.773	0.795	0.810	0.826	0.848	0.948	0.810	0.023	0.271
	NPV	0.555	0.709	0.733	0.750	0.767	0.791	0.908	0.750	0.025	0.354
	AUROC	0.757	0.823	0.843	0.857	0.870	0.890	0.928	0.857	0.020	0.171
Artificial Neural Network	Accuracy	0.689	0.742	0.765	0.781	0.797	0.820	0.877	0.781	0.024	0.188
	F1	0.677	0.735	0.758	0.774	0.791	0.814	0.888	0.774	0.024	0.211
	Sensitivity	0.672	0.757	0.780	0.796	0.811	0.834	0.884	0.796	0.023	0.212
	Specificity	0.591	0.728	0.752	0.768	0.785	0.808	0.928	0.768	0.024	0.337
	PPV	0.659	0.771	0.793	0.808	0.824	0.846	0.940	0.808	0.023	0.281
	NPV	0.550	0.708	0.732	0.749	0.766	0.790	0.912	0.749	0.025	0.361
	AUROC	0.751	0.812	0.833	0.847	0.861	0.881	0.949	0.847	0.021	0.198
Adaptive Boosting	Accuracy	0.683	0.736	0.759	0.775	0.791	0.815	0.885	0.775	0.024	0.202
	F1	0.674	0.735	0.758	0.775	0.791	0.814	0.890	0.775	0.024	0.216
	Sensitivity	0.671	0.759	0.781	0.797	0.813	0.835	0.889	0.797	0.023	0.217
	Specificity	0.585	0.732	0.756	0.772	0.789	0.812	0.941	0.772	0.024	0.356
	PPV	0.676	0.780	0.802	0.817	0.832	0.853	0.951	0.817	0.022	0.274
	NPV	0.567	0.709	0.733	0.750	0.767	0.791	0.929	0.750	0.025	0.362
	AUROC	0.755	0.829	0.848	0.862	0.875	0.895	0.929	0.862	0.020	0.175

Note: Summary of model metrics for each of the four machine learning techniques (XGBoost, Random Forest, Artificial Neural Network, and Adaptive Boosting) based upon the derived distribution using analytic formulas described within the study.

Abbreviations: AUROC, area under the receiver operator characteristic curve; NPV, negative predictive value; PPV, positive predictive value.