RLMD-PA: A Reinforcement Learning-Based Myocarditis Diagnosis Combined with a Population-Based Algorithm for Pretraining Weights

Table 4

5-CV classification performances (accuracy, recall, and precision) obtained for automated myocarditis detection using various machine learning algorithms with the Z-Alizadeh Sani myocarditis dataset.

	Accuracy					Recall					Precision
Method	Min	Median	Max	Mean	Std.dev.	Min	Median	Max	Mean	Std.dev.	Min	Median	Max	Mean	Std.dev.
SVM	0.568	0.691	0.754	0.683	0.070	0.674	0.745	0.778	0.737	0.042	0.450	0.565	0.651	0.565	0.074
KNN	0.480	0.614	0.635	0.588	0.064	0.399	0.637	0.683	0.589	0.111	0.337	0.490	0.511	0.460	0.072
Naïve Bayes	0.547	0.632	0.676	0.615	0.051	0.388	0.534	0.713	0.565	0.134	0.395	0.510	0.553	0.484	0.062
Logistic regression	0.627	0.662	0.720	0.661	0.038	0.583	0.658	0.741	0.657	0.057	0.503	0.542	0.603	0.541	0.041
Random forests	0.415	0.550	0.590	0.530	0.070	0.537	0.683	0.711	0.648	0.071	0.329	0.437	0.469	0.420	0.056