Comparing the Prognostic Value of Stress Myocardial Perfusion Imaging by Conventional and Cadmium-Zinc Telluride Single-Photon Emission Computed Tomography through a Machine Learning Approach

Table 4

Machine learning analysis and statistical comparison through chi square test for proportions after SMOTE implementation.

		Accuracy (%)	Error (%)	Recall (%)	Specificity (%)
Tree	C-SPECT	88.1	11.9	86.2	90.1
	CZT-SPECT	88.1	11.9	86.9	89.3
	value	1.000		0.760	0.731
KNN	C-SPECT	91.9	8.1	83.9	99.8
	CZT-SPECT	91.6	8.4	84.7	98.5
	value	0.858		0.774	0.058
SVM	C-SPECT	91,5	8.5	87,6	95.0
	CZT-SPECT	94.5	5.5	92.2	96.8
	value	0.016		0.028	0.279
NB	C-SPECT	59.3	40.7	86.7	32.0
	CZT-SPECT	59.0	41.0	87.6	30.3
	value	0.880		0.677	0.599
RF	C-SPECT	93.4	6.6	90.3	94.4
	CZT-SPECT	93.0	7.0	91.0	94.9
	value	0.637		0.720	0.757

Abbreviations. KNN: nearest neighbor; SVM: support vector machine; NB: Naïve Bayes; RF: random forests.