A Comparison among Different Machine Learning Pretest Approaches to Predict Stress-Induced Ischemia at PET/CT Myocardial Perfusion Imaging

Table 1

Clinical characteristics of cohort according to MPI outcome.

	Normal ()	Ischemic ()	value
Age, (%)			<0.001
<55	777 (39)	84 (17)
55-65	603 (30)	146 (29)
>65	622 (31)	271 (54)
Male gender, (%)	881 (44)	334 (67)	<0.001
Body mass index ≥30, (%)	1024 (51)	258 (52)	0.93
Chest pain, (%)			<0.001
Typical	678 (34)	114 (23)
Atypical	256 (13)	87 (17)
Noncardiac	1068 (53)	300 (60)
Diabetes, (%)	479 (24)	187 (37)	<0.001
Dyspnea, (%)	446 (22)	139 (28)	<0.05
Family history of CAD, (%)	945 (47)	199 (40)	<0.005
Hypertension, (%)	1361 (68)	401 (80)	<0.005
Hyperlipidemia, (%)	1210 (60)	343 (69)	<0.005
Smoking, (%)	557 (28)	144 (29)	0.72
Diagnostic question, (%)^§			<0.001
Diagnostic evaluation	1642 (82)	370 (74)
Presurgery evaluation	360 (18)	131 (26)

Considering noncardiac patients as the reference. ^§Considering diagnostic evaluation patients as the reference.