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

Table 2

Values used for tuning of parameters for each ML technique.

	Parameter	Parameter space	Chosen value
ADA	Number of trees	10, 25, 50, 100, 200	25
	Max tree depth	5, 10, 20, 50	10
	Learning rate	0.001, 0.005, 0.01, 0.05, 0.1, 0.5	0.01
AdaBoost	Number of trees	10, 25, 50, 100, 200	50
	Method	AdaBoost.M1, real AdaBoost	AdaBoost.M1
Logistic	Family	Binomial	Binomial
Naïve Bayes	Laplace correction	0, 0.5, 1.0	0
	Distribution type (kernel)	True, false	False
	Bandwidth adjustment	0.01, 0.05, 0.1, 0.5, 1.0	0.1
Random Forest	Number of randomly selected predictors	3, 5, 10, 20	10
Rpart	Minimum number of observations in a node	10, 15, 30	15
	Minimum number of observations in any leaf node	3, 5, 10	5
	Max tree depth	3, 5, 10, 20	10
	Complexity parameter of the tree	0.0001, 0.001, 0.01, 0.1	0.001
SVM	Kernel	Linear, radial, sigmoid	Sigmoid
	Parameter needed for sigmoid	0.05, 0.1, 0.25, 0.5	0.1
	Cost	0.5, 1, 2, 5	1
XGBoost	Number of trees	25, 50, 100, 200	100
	Max tree depth	5, 10, 20	10
	Learning rate	0.001, 0.005, 0.01, 0.05, 0.1, 0.5	0.01
	Subsamples	0.5, 0.75, 1	1