Table 1.

Classification performance of the supervised models with the classification threshold set at 50%.

		Classification performance (%)
Model and metrics		Ca vs Bb	C vs NBc	B vs NB	C vs ALSd
Random forest
	Accuracy	93.6	91.1	75.5	90.3
	Sensitivity	91.1	92.1	55.7	92.1
	Specificity	95.5	89.6	88.4	85.7
Naïve Bayes
	Accuracy	91.0	87.9	75.4	90.3
	Sensitivity	89.2	86.7	62.7	92.1
	Specificity	93.2	90.0	81.2	85.7
Logistic regression
	Accuracy	93.8	91.4	70.1	91.1
	Sensitivity	92.5	89.1	62.2	89.6
	Specificity	94.8	95.6	73.5	93.3
Linear discriminant analysis
	Accuracy	94.3	91.6	71.2	91.6
	Sensitivity	95.6	87.4	61.8	88.3
	Specificity	90.0	98.8	75.4	87.8
Neuronal network
	Accuracy	94.8	92.5	70.4	92.2
	Sensitivity	91.7	90.3	60.0	90.8
	Specificity	97.2	96.4	75.2	95.6
Support vector machine
	Accuracy	95.8	91.5	69.9	91.6
	Sensitivity	91.4	88.4	59.4	88.9
	Specificity	99.3	97.0	74.6	98.2

^aC: control group.

^bB: patients with amyotrophic lateral sclerosis (ALS) with bulbar involvement.

^cNB: patients with ALS without bulbar involvement.

^dALS: all patients with ALS.