Juliane Budde

Ultra-high field MRI systems of 7 T and above have been emerging within the last years with the goal of taking advantage of the expected increase in signal-to-noise ratio (SNR) and possibly unique contrast mechanisms. These developments are intricately tied with a number of specific problems not encountered at conventional fields. These effects, however, also open up new possibilities for novel and improved image contrast, permitting better visualization of internal structures as well as the venous vessel network with susceptibility weighted imaging (SWI).

One of the major goals in functional imaging research is to determine and optimized the spatial specificity of the BOLD response. Combining fMRI with high resolution anatomical information can help to localize and eliminate venous signal contributions (2,3). This is further enhanced by the higher SNR and predicted increase in BOLD signal expected at higher magnetic field strength, allowing better spatial resolution, as well as by the improved spatial specificity that is expected especially for spin-echo fMRI. Furthermore, phase imaging, which can be capable of detecting structural variations within the cortex, can help to distinguish different parts of the cortex according to its morphology improving between-subject comparison or can also aid to gain layer-specific activation. Thus, human functional brain imaging in normal subjects as well as in patients can benefit in terms of specificity and sensitivity.

Anatomical and Functional Human Imaging at 9.4 T

Introduction

With increasing field strength, the MR signal-to-noise ratio is expected to grow linearly, while a BOLD increase of more than linear is expected. In addition, anatomical imaging also profits from increased venous contrast at high field strength. For functional imaging, a higher emphasis on signal from microvasculature is predicted, especially for SE-EPI, while GE-EPI remains more specific to the macrovasculature

Goals

The improved venous contrast and high SNR is to be used to acquire very high resolution T₂^*- weighted images, from which phase images and susceptibility images can be derived. The feasibility of GRE-EPI and SE-EPI studies at 9.4 T should be examined, and optimal parameter settings should be determined to enable routine scanning at this field strength. Additionally, the contributions to the functional signal from GRE-EPI and SE-EPI originating from micro- and macrovasculature are to be determined.

Methods

SE-EPI as well as GRE-EPI images are acquired with additional co-localized high resolution T₂^* weighted GRE images. Phase and susceptibility-weighted images are calculated [2]. The functional maps are overlaid on T₂^*-weighted and phase images. Co-registration of both image modalities can demonstrate where the functional signal coincides with venous structures.

Initial results

T₂^*-weighted and susceptibility-weighted images with a resolution of 1.75 mm x 1.75 mm x 1.3 mm reveal venous vasculature at high detail, phase images show excellent gray and white matter contrast. GRE-EPI images at a resolution of 1 mm isotropic show robust activation with a finger-tapping paradigm.  Activation size is around 8 % outside large veins and around 12 % inside large veins [3].

Initial conclusion

High resolution anatomical images profits from ultra-high field in terms of enhanced T₂^* contrast and increased SNR. GRE-EPI is feasible at resolution of 1 mm, and shows higher percent changes in macrovasculature.

Figure: EPI and anatomical images are shown with overlaid activation and location of veins. Activation is shown in red -yellow color-scale, veins are delineated in light blue. Location of veins was determined from T2*-weighted images (c, d) and registered onto EPI images (b). Figure a) and b) are EPI, T2*-weighted images are shown in c) and d). Phase images e) and f) depict gray-white matter contrast.

References

1.      Uludag K, Müller-Bierl B and Ugurbil K (2009) An integrative model for neuronal activity-induced signal changes for gradient and spin echo functional imaging NeuroImage 48(1) 150-165.

2.      Budde J, Shajan G, Hoffmann J, Ugurbil K and Pohmann R (2011) Human imaging at 9.4 T using T2*-, phase-, and susceptibility-weighted contrast Magnetic Resonance in Medicine 65(2) 544-550.

3.      Budde J, Mühlbauer F, Gunamony S, Zaitsev M and Pohmann R (2011): Human fMRI at 9.4 T: Preliminary Results, 19th Annual Meeting and Exhibition of the International Society for Magnetic Resonance in Medicine (ISMRM 2011), 19(1636).