Using predictive process monitoring to assist thrombolytic therapy decision-making for ischemic stroke patients

Table 5 Experimental results on test set

Clinical application	Prediction model	ACC	Precision		Recall		F1		AUC
Clinical application	Prediction model	ACC	throm	Non-throm	throm	Non-throm	throm	Non-throm	AUC
Intravenous thrombolysis	DT without control flow	0.87	0.49	0.96	0.77	0.89	0.60	0.92	0.83
	DT with control flow	0.93	0.64	0.99	0.96	0.92	0.77	0.96	0.94
	RF without control flow	0.88	0.51	0.99	0.93	0.87	0.65	0.93	0.90
	RF with control flow	0.96	0.82	0.98	0.88	0.97	0.85	0.98	0.93
	Clinical guidelines	0.91	0.64	0.95	0.68	0.95	0.66	0.95	0.81
Arterial thrombolysis	DT without control flow	0.74	0.29	0.93	0.62	0.76	0.39	0.83	0.69
	DT with control flow	0.85	0.48	0.97	0.82	0.86	0.60	0.91	0.84
	RF without control flow	0.82	0.39	0.92	0.54	0.86	0.45	0.89	0.70
	RF with control flow	0.91	0.65	0.96	0.76	0.93	0.70	0.95	0.85
	Clinical guidelines	0.89	0.73	0.91	0.36	0.98	0.48	0.94	0.67

ISSN: 1472-6947