Table 6 Neural network prediction using A) CSF data and B) serum data.

From: Metallomic Biomarkers in Cerebrospinal fluid and Serum in patients with Parkinson’s disease in Indian population

A)	BayesNet	NaiveBayes	SimpleLogistic	RBFNetwork	JRip	Random Forest	Multilayer Perceptron
CSF Class Prediction (Normal and Parkinson’s disease)
Accuracy	99.78	98.18	96.72	98.63	94.54	98.63	97.45
Precision	1	0.9	1	0.9	0.9	0.9	1
Recall	0.9	0.9	0.9	0.9	0.9	0.9	0.9
F-measure	0.9	0.9	0.9	0.9	0.9	0.9	0.9
Area under ROC	1	0.9	0.9	0.9	0.9	0.9	0.9
CSF Stages Prediction (normal, stage1, stage2 and stage3 PD)
Accuracy	71.73	68.64	70.73	69.36	67.18	70.36	68.91
Precision	0.6	0.4	0.5	0.3	0.4	0.5	0.4
Recall	0.7	0.5	0.5	0.6	0.4	0.6	0.5
F-measure	0.7	0.4	0.5	0.4	0.4	0.5	0.4
Area under ROC	0.8	0.8	0.8	0.7	0.6	0.8	0.8
CSF Status Prediction (Normal, Progressive and Static)
Accuracy	79.72	82.45	83.72	83.27	80.90	84.09	82.27
Precision	0.7	0.7	0.7	0.7	0.7	0.7	0.7
Recall	0.7	0.6	0.8	0.8	0.7	0.8	0.7
F-measure	0.7	0.7	0.7	0.7	0.74	0.7	0.7
Area under ROC	0.9	0.9	0.9	0.8	0.8	0.9	0.9
B) Serum Class Prediction (Normal and Parkinson’s disease)
Accuracy	99.87	98.63	99.67	98.80	94.54	96.72	99.78
Precision	0.9	0.9	0.9	0.9	0.9	1	0.9
Recall	1	0.9	1	1	0.9	0.9	1
F-measure	0.9	0.9	0.9	0.9	0.9	0.9	0.9
Area under ROC	1	0.9	1	0.9	0.9	0.9	1
Serum Stages Prediction (normal, stage1, stage2 and stage3 PD)
Accuracy	65.06	69.28	68.46	69.16	72.95	67.33	68.29
Precision	0.7	1	0.9	0.9	1	0.9	0.9
Recall	0.9	1	0.9	1	1	0.9	1
F-measure	0.8	1	0.9	0.9	1	0.9	0.9
Area under ROC	0.8	1	1	0.9	1	0.9	0.9
Serum Status Prediction (Normal, Progressive and Static)
Accuracy	88.64	87.91	90.57	88.89	87.95	88.56	91.07
Precision	0.8	1	0.9	0.9	0.9	0.9	0.9
Recall	0.8	0.9	1	1	0.9	0.9	1
F-measure	0.7	0.9	0.9	0.9	0.9	0.9	0.9
Area under ROC	0.8	0.8	0.9	0.9	0.9	0.9	0.9

Back to article page

Quick links

Search

Quick links