A machine learning model to predict critical care outcomes in patient with chest pain visiting the emergency department

Table 5 Performance of LASSO and comparison models in predicting the critical care outcomes in patients with chest pain

	Accuracy	Sensitivity	Specificity	PPV/Precision	NPV	F1	Cut-off	AUC	95%CI
Training set
LASSO	0.861	0.847	0.881	0.853	0.867	0.845	113.0	0.924	0.896–0.952
HEART	0.654	0.839	0.476	0.578	0.815	0.694	5.5	0.699	0.644–0.754
GRACE	0.707	0.588	0.805	0.714	0.703	0.645	145.5	0.737	0.684–0.791
TIMI	0.665	0.471	0.826	0.692	0.652	0.560	5.5	0.701	0.646–0.756
Testing set
LASSO	0.890	0.864	0.911	0.891	0.889	0.877	117.0	0.953	0.922–0.984
HEART	0.710	0.773	0.658	0.654	0.776	0.708	6.5	0.754	0.675–0.832
GRACE	0.717	0.606	0.810	0.727	0.711	0.661	141.5	0.747	0.664–0.829
TIMI	0.683	0.500	0.835	0.717	0.667	0.589	4.5	0.735	0.655–0.815

PPV positive predictive value, NPV negative predictive value, AUC area under the receiver-operating-characteristics curve

ISSN: 1471-227X