Chapter 5
EEG Source Imaging and Multimodal
Neuroimaging
Yingchun Zhang
Abstract During most common applications, the signals produced by cortical
dipoles are detected at the scalp level. While these measurements can be highly
informative and feature optimal temporal resolution, their spatial detection is hindered by the conduction through the tissues of the head. A deeper understanding
of cortical sources can be provided by source imaging techniques. These methods
first create mathematical models of the head, assigning appropriate properties to
each layer. Brain activity is then calculated based on these assigned models and
the observed EEG measurements, greatly improving the spatial resolution of EEG
measurement and providing insights regarding otherwise hidden cortical dynamics.
Source Imaging approaches can be further enhanced by integrating a second imaging
modality. This is particularly useful with imaging methods that feature high spatial
resolution or whose signals are not blurred by transduction. In the following chapter,
we provide a detailed introduction to the general principles and basic algorithms
of source imaging techniques. The discussion then expands to explore how other
modalities can interact with these techniques to improve our results. At its conclusion, readers should have a good idea on how EEG data can be expanded to provide
cortical insight.
After addressing the complexities of signal processing and analysis, the observation
of EEG and ERP signals has been relatively straightforward. The conductive nature
of these electrical signals opens up an intriguing new possibility for examining brain
activity; using mathematical approaches to determine the most likely cortical sources
of scalp potentials. Performing this backwards calculation is known as electrical
source imaging (ESI). ESI techniques then present us with appealing imaging properties by allowing us to more directly observe brain activity while maintaining the
low cost, high efficiency, and noninvasive features of EEG.
Before embarking on a detailed discussion of the algorithms and methods used
in source imaging, it’s first necessary to understand the cells and tissues within the
Y. Zhang (B)
Department of Biomedical Engineering, University of Houston, Houston, USA
e-mail: yzhang94@uh.edu
© Springer Nature Singapore Pte Ltd. 2018
C.-H. Im (ed.), Computational EEG Analysis, Biological and Medical Physics,
Biomedical Engineering, https://doi.org/10.1007/978-981-13-0908-3_5
83
EEG Source Imaging and Multimodal
Neuroimaging
Yingchun Zhang
Abstract During most common applications, the signals produced by cortical
dipoles are detected at the scalp level. While these measurements can be highly
informative and feature optimal temporal resolution, their spatial detection is hindered by the conduction through the tissues of the head. A deeper understanding
of cortical sources can be provided by source imaging techniques. These methods
first create mathematical models of the head, assigning appropriate properties to
each layer. Brain activity is then calculated based on these assigned models and
the observed EEG measurements, greatly improving the spatial resolution of EEG
measurement and providing insights regarding otherwise hidden cortical dynamics.
Source Imaging approaches can be further enhanced by integrating a second imaging
modality. This is particularly useful with imaging methods that feature high spatial
resolution or whose signals are not blurred by transduction. In the following chapter,
we provide a detailed introduction to the general principles and basic algorithms
of source imaging techniques. The discussion then expands to explore how other
modalities can interact with these techniques to improve our results. At its conclusion, readers should have a good idea on how EEG data can be expanded to provide
cortical insight.
After addressing the complexities of signal processing and analysis, the observation
of EEG and ERP signals has been relatively straightforward. The conductive nature
of these electrical signals opens up an intriguing new possibility for examining brain
activity; using mathematical approaches to determine the most likely cortical sources
of scalp potentials. Performing this backwards calculation is known as electrical
source imaging (ESI). ESI techniques then present us with appealing imaging properties by allowing us to more directly observe brain activity while maintaining the
low cost, high efficiency, and noninvasive features of EEG.
Before embarking on a detailed discussion of the algorithms and methods used
in source imaging, it’s first necessary to understand the cells and tissues within the
Y. Zhang (B)
Department of Biomedical Engineering, University of Houston, Houston, USA
e-mail: yzhang94@uh.edu
© Springer Nature Singapore Pte Ltd. 2018
C.-H. Im (ed.), Computational EEG Analysis, Biological and Medical Physics,
Biomedical Engineering, https://doi.org/10.1007/978-981-13-0908-3_5
83