Linear and non-linear effects of gradient artifact filtering methods in simultaneous EEG-FMRI

Monica Cusenza, Agostino Accardo, Fabrizio Monti, Placido Bramanti

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Simultaneous EEG-fMRI is a powerful emerging tool in functional neuroimaging that exploits the relationship between neuronal electrophysiological activity and its hemodynamic response. It has found application in the study of both spontaneous and evoked brain activity. Combining the complementary advantages of the two techniques it provides a measurement with high temporal and spatial resolution, allowing a reliable localization of event generators. However, EEG data recorded inside MRI scanner are heavily corrupted by different types of artifacts due to the interactions between the patient, EEG electrodes wires and the magnetic fields inside the scanner. In particular, gradient switching and RF pulses, necessary to acquire fMRI data, generate large artifacts that can completely obscure EEG signals. Many methods have been proposed to eliminate or at least reduce gradient artifact. In this paper both a qualitative and a quantitative evaluation of two different algorithms used for gradient artifact removal are presented. Linear and non-linear characteristics of EEG, such as power spectra, fractal dimension and beta scaling exponent, are evaluated for EEGs recorded outside and inside the scanner, in MR static and dynamic conditions. The study highlights how residual artifacts after correction and artifacts induced by correction itself could still considerably affect EEG signals. The results suggest that the quality of both these gradient artifact filtering methods is not yet sufficient to preserve EEG characteristics and thus it must be further improved. The aim of this study is to make neurophysiologists aware of the filtering effects that can compromise linear and non-linear analysis of EEG recorded during functional MRI.

Original languageEnglish
Title of host publication47th Annual Rocky Mountain Bioengineering Symposium and 47th International ISA Biomedical Sciences Instrumentation Symposium 2010
Pages282-287
Number of pages6
Volume480
Publication statusPublished - 2010
Event47th Annual Rocky Mountain Bioengineering Symposium and 47th International ISA Biomedical Sciences Instrumentation Symposium 2010 - Laramie, WY, United States
Duration: Apr 9 2010Apr 11 2010

Other

Other47th Annual Rocky Mountain Bioengineering Symposium and 47th International ISA Biomedical Sciences Instrumentation Symposium 2010
CountryUnited States
CityLaramie, WY
Period4/9/104/11/10

Keywords

  • EEG
  • fMRI
  • Fractal dimension
  • Gradient artifact removal
  • Power spectra
  • Simultaneous recording

ASJC Scopus subject areas

  • Bioengineering
  • Biomedical Engineering

Fingerprint Dive into the research topics of 'Linear and non-linear effects of gradient artifact filtering methods in simultaneous EEG-FMRI'. Together they form a unique fingerprint.

  • Cite this

    Cusenza, M., Accardo, A., Monti, F., & Bramanti, P. (2010). Linear and non-linear effects of gradient artifact filtering methods in simultaneous EEG-FMRI. In 47th Annual Rocky Mountain Bioengineering Symposium and 47th International ISA Biomedical Sciences Instrumentation Symposium 2010 (Vol. 480, pp. 282-287)