A new method for quantifying mfVEP signal intensity in multiple sclerosis. (September 2015)

• Record Type: Journal Article
• Authors: Castillo, M. Ortiz del; de Santiago, L.; Fernández, A.; Blanco, R.; Rodríguez-Ascariz, J.M.; Barea, R.; Miguel-Jiménez, J.M.; Sánchez-Morla, E.M.; Boquete, L.
• Is Part Of: Biomedical signal processing and control. Volume 22(2015)
• Page Start: 119
• Page End: 125
• Publication Date: 2015-09
• Subjects: Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2015.06.015
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: The paper proposes the Percentage of Energy (PoE) method to evaluate mfVEP response intensity. The PoE parameter clearly characterizes mfVEP signals' amplitude. The PoE parameter produces better index of discrimination between controls and patients. The PoE parameter achieves better test–retest reliability. Abstract: The multifocal visual evoked potential (mfVEP) test measures the potentials obtained by simultaneous stimulation of multiple areas of the visual field. The principal mfVEP signal characteristics analysed are intensity and latency. A new parameter named Percentage of Energy (PoE) based on energy distribution along the recording is defined to evaluate (mfVEP) response intensity. mfVEP signals from controls and subjects classified according to the risk of developed Multiple Scleroris (MS) are used: 24 control, 15 Radiologically Isolated Syndrome (RIS), 28 Clinically Isolated Syndrome (CIS) and 28 MS diagnosed. Eyes from CIS and MS groups have been classified in neuritis optic eyes (NO) and no affected by neuritis optics eyes (nON). mfVEP signals intensity are analysed using the classic signal-to-noise (SNR) ratio and the PoE parameter. Index on discrimination between controls and different groups of diagnosed patients based on area under the curve are calculated using SNR and PoE. Lastly reproducibility test of SNR and PoE are analysed. The mean values of coefficient of variability ( C v ) for all the subjects are C VSRN = 0.34 and C VPOE = 0.17 ( p < 0.05). Significance difference between diagnosed groups is found in eight cases if SNR is used to quantify intensity of mfVEP signals. However, if PoE parameter is used, significance difference is founded in 10 cases. The mean area under the curve between control and diagnosed patients is increased from AUCSNR = 0.79 to AUCPOE = 0.83 ( p < 0.05). Better concordance in the test–retest measurements is obtained by using the PoE parameter. In conclusion, the PoE parameter presents advantages over the SNR parameter when characterizing mfVEP signal amplitude.