Skip to main content

View full-text article in PMC

. 2019 Oct 30;7(10):e14926. doi: 10.2196/14926

Table 2.

Test performance of the abnormal blood pressure data analysis algorithm compared to other models.

Risk levels, models		F1-score	Specificity	Accuracy	Precision	Recall	AUC^a
Low
	FCNN^b	0.645	0.867	0.835	0.588	0.714	0.863
	RF^c	0.552	0.702	0.725	0.419	0.809	0.859
	ABA^d	0.659	0.867	0.840	0.596	0.738	0.904
Caution
	FCNN	0.771	0.628	0.710	0.790	0.753	0.731
	RF	0.571	0.785	0.565	0.794	0.446	0.636
	ABA	0.786	0.629	0.725	0.795	0.776	0.756
High
	FCNN	0.528	0.936	0.875	0.560	0.500	0.827
	RF	0.519	0.848	0.83	0.434	0.646	0.894
	ABA	0.673	0.953	0.885	0.618	0.714	0.912

^aAUC: area under the curve.

^bFCNN: fully connected neural networks.

^cRF: random forest.

^dABA: abnormal blood pressure data analysis.