Table 4 Comparison of classification performance with existing research.

References	Sample size	Input feature	Feature selection method	Classifier	Cross validation method	Performance
Shen et al.⁴	SZ = 64 HC = 53	Gut microbiota	Boruta variable selection	RF	None	AUC = 0.837
Brisa et al.¹⁴	SZ = 58 HC = 123	Blood and cognitive	PLS-DA	LDA	Tenfold	Accuracy = 0.86 AUC = 0.89
Jason et al.²⁰	SZ = 40 HC = 12	EEG	None	SVM	None	Accuracy = 0.87 Sensitivity = 0.90 Specificity = 0.77
Sai Krishna Tikka et al.²¹	SZ = 38 HC = 20	EEG	None	SVM	Hold-out	Accuracy = 0.79 Sensitivity = 0.92 Specificity = 0.50 AUC = 0.71
Our best	SZ = 49 HC = 50	Gut microbiota Blood EEG	None	SVM	Tenfold	Accuracy = 0.92 Sensitivity = 0.92 Specificity = 0.92 AUC = 0.97

RF random forest, PLS-DA partial least squares discriminant analysis, LDA linear discriminant analysis, EEG electroencephalogram, SVM support vector machine, AUC area under the receiver operating characteristic curve.

Quick links

Search