| Budget Amount *help |
¥3,770,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥270,000)
Fiscal Year 2007: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2006: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2005: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2004: ¥800,000 (Direct Cost: ¥800,000)
|
|---|
| Research Abstract |
The target of this investigation is to finally realize the ultimate promise of electromagnetic brain imaging, which is the spatiotemporal localization of cortical networks involved in sensorimotor, language, memory, and higher cognitive functions. We first compare various adaptive and non-adaptive spatial filters in this context Two important properties for any inverse algorithm are the localization bias and the spatial resolution. We discussed these properties for various adaptive and non-adaptive spatial filters. We also conducted arguments on the signal-to-noise ratio (SNR)of the spatial filter output and discusses the degradation of the output SNR caused by errors in the forward-field calculations, and presents methods that are robust to these errors. We then discuss the effects of external interference on the performance of adaptive spatial filters.
The major problem with brain electromagnetic measurements is that the measured signal generally contains a large amount of interfering
… More
magnetic fields. We deal with interference of a physiological origin. In particular, we focuses on the background brain activity, which is sometimes referred to as brain noise or physiological noise. The prominent characteristic of the background interference is that it is generated by a large number of sources. Hence, the underlying low-rank signal assumption, which is necessary for formulating adaptive spatial filters, is invalidated. We present a theoretical analysis showing that such high-rank interference can cause a severe spatial blur in the reconstruction results. We then propose a prewhitening eigenspace-projection spatial filter, which can achieve the source reconstruction free from the influence of background interference even when the nnwer of the interference is significantly large.
We then proceed with a study on mapping functional connectivity and an investigation toward this direction is mapping of mean imaginary coherence. In this investigation, resting-state MEG was recorded from a patient with brain lesions. The hypothesis here is that such a patient has decreased connectivity around pathologic regions, and such decreased connectivity can be detected by mapping the mean imaginary coherence. When applying the method to MEG from a patient with brain lesions, voxels showing decreased mean imaginary coherence are found near the brain tumor and thus we conclude that mapping mean imaginary coherence can provide useful clinical information on brain lesions. We are now conducting systematic investigations for evaluating the usefulness of this mean imaginary coherence mapping as a new clinical tool. Less
|
