A New Method of Tracking of WM Crossing Fiber Bundles Based on QBI

doi:10.4236/eng.2013.510B051

Paper Menu >>

Journal Menu >>

Engineering, 2013, 5, 250-251

http://dx.doi.org/10.4236/eng.2013.510B051 Published Online October 2013 (http://www.scirp.org/journal/eng)

A New Method of Tracking of WM Crossing Fiber

Bundles Based on QBI*

Zhanxiong Wu

College of Electronic Information, Hangzhou Dianzi University, Hangzhou, China

Email: [email protected]

Received June 2013

ABSTRACT

Tracking of crossing WM fiber bundles can be resolved using diffusion MRI imaging. DTI can only resolve a single

fiber orientation within each voxel due to the constraints of the tensor model. DSI requires large pulse d field gradients

and time-intensive sampling. This paper puts forward with a new method based on QBI, which uses a spherical tomo-

graphic invers ion called Funk-Radon tran sform to get high angular resolution diffusion imaging signal. From the track-

ing results, we can get the conclusion that QBI-tracking can resolve crossing fiber time-savingly.

Keywords: Diffusio n Tensor Imaging; Diffusion Spectrum Imaging; q-Ball

1. Introduction

White matter is mainly composed of nerve fibers, which

are the connected channels of brain different function

areas. The tracking of nerve fibers is very important for

white matter disease, and is of great significance for cog-

nitive research. Diffusion MRI has been widely used to

assess the integrity of axonal fibers because of its unique

ability to map fiber orientations in vivo.

Diffusion tensor imaging (DTI) provides a powerful

tool for mapping neural histoarchitecture in vivo. How-

ever, DTI can only resolve a single fiber orientation

within each imaging voxel due to the constraints of the

tensor model. DTI cannot resolve fibers crossing, bend-

ing, or twisting within an individual voxel [1-4].

As to diffusion spectrum imaging (DSI), diffusion is

described with the probability density function (PDF)

which for each voxel specifies the 3D distribution of mi-

croscopic displacements of MR-visible spins that it con-

tains. Reconstruction of the diffusion PDF by Fourier

transform of the diffusion signal forms the basis of the

QSI method. But DSI requires large pulsed field gra-

dients and time-intensive sampling [5-7].

Q-ball imaging can resolve multiple orientations of

crossing fiber in one voxel and does not require any as-

sumptions on the di ffusion p r oc e s s o f water m olecules.

2. Method and Material

2.1. Reconstruction of the Diffusion Orientation

Distribution Function (ODF)

QBI is reconstructed based on the relationship of the in-

terested ODF vector and its orthogonal plane projected

on to the acquired q-space data. The ODF was directly

calculated from the attenuated echo signal on a shell in

the q-space with a fixed b-value based on the Funk-Ra-

don transform approach.

ODF()=( ,)d

Eq q

⊥

∆

∫

(1)

where u is the unit vector for the desired ODF direction,

and C is the normalization constant, and E is the atte-

nuated echo signal [8].

The diffusion MRI dataset was provided by Advanced

Biomedical MRI Lab of National Taiwan University

Hospital. The acquisition equipment was 3 Tesla Trio of

the company Siemens of Germany, 1.9 mm × 1.9 mm ×

1.9 mm, TR = 11,500 ms, TE = 31 ms, △/δ = 80/35 ms,

the maximal b value is 4000 s/mm2. Figure 1 shows the

reconstruction of QBI.

2.2. Tracking of Crossing Fiber Bundles

Firstly all of local maximums of ODF were got, then the

parameters of angle threshold and stepping size were set;

At last we found the smooth lines among the adjacent

voxels. Here the value of angle threshold was set to 60

degrees, and the value of stepping size was set to 1.4 mm.

The constraint of fiber bundle length was between 16

mm and 86 mm.

3. Results

Figure 2 is the tracking of whole brain WM. The first

low is based on DTI, the second one is based on DSI, and

the third one is based on QBI. Figure 3 is the tracking of

*The research is supported by NFSC-51207038 and ZJFSC-Q12F010017.

Z. X. WU

251

Figure 1. Reconstruction of QBI, (a) Sagittal view; (b) Co-

ronal view; (c) Axial view.

Figure 2. Tracking of whole brain. (a)-(c) DTI; (d)-(f) DSI;

(g)-(i) QBI.

Figure 3. Tracking of corpus callosum. (a) DTI; (b) DSI; (c)

QBI.

corpus callosum.

4. Conclusion

The ability of QBI to resolve crossing fibers is of great

significance that helps to understand the detailed archi-

tecture of the cerebral white matter. Compared to DTI

and DSI, the tracking based on QBI is capable of distin-

guishing crossing fiber bundles within each voxel time-

savingly.

REFERENCES

[1] D. Bihan, J. F. Mangin and C. Poupon, “Diffusion Tensor

Imaging: Concepts and Applications,” Journal of Mag-

netic Resonance Imaging, Vol. 13, No. 4, 2001, pp. 534-

546. http://dx.doi.org/10.1002/jmri.1076

[2] M .Wiegell, D. Tuch and H. Larson, “Automatic Seg-

mentation of Thalamic Nuclei from diffusion Tensor

Magnetic Resonance Imaging,” NeuroImage, Vol. 19,

2003, pp. 391-402.

http://dx.doi.org/10.1016/S1053-8119(03)00044-2

[3] P. J. Basser, J. Mattiello and D. LeBihan, “MR Diffusion

Tensor Spectroscopy and Imaging,” Biophysical Journal,

Vol. 66, 1994, pp. 259-267.

http://dx.doi.org/10.1016/S0006-3495(94)80775-1

[4] P. Hagmann, L. Jonasson and P. Maeder, “Understanding

Diffusion MR Imaging Techniques: From Scalar Diffu-

sion-Weighted Imaging to Diffusion Tensor Imaging and

Beyond,” Radio Graphics, Vol. 26, 2006, pp. 205-223.

http://dx.doi.org/10.1148/rg.26si065510

[5] G. K. Deutsch, R. F. Dougherty and R. B ammer, “Child-

ren ’s Reading Performance Is Correlated with White

Matter Structure Measured by Diffusion Tensor Imaging,”

Cortex, Vol. 41, No. 3, 2005, pp. 354-363.

http://dx.doi.org/10.1016/S0010-9452(08)70272-7

[6] S. Mori and P. C. Zij l, “Fiber Tracking: Principles and

Strategies-Atechnical Review,” NMR in Biomedicine, Vol.

15, 2002, pp. 468-480. http://dx.doi.org/10.1002/nbm.781

[7] V. J. Wedeen, P. Hagmann and W. Y. Tseng, “Mapping

Complex Tissue Architecture with Diffusion Spectrum

Magnetic Resonance Imaging,” Magnetic Resonance in

Medicine, Vol. 54, 2005, pp. 1377-1386.

http://dx.doi.org/10.1002/mrm.20642

[8] J. S. Campbell, K. Siddiqi and V. V. Rymar, “Flow-

Based Fiber Tracking with Diffusion Tensor and q-Ball

Data: Validation and Comparison to Principal Diffusion

Direction Techniques,” Neuroimage, Vol. 27, No. 4, 2005,

pp. 725-736.

http://dx.doi.org/10.1016/j.neuroimage.2005.05.014