This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Introduction: Line scan diffusion imaging (LSDI) has been shown to be relatively insensitive to motion artifacts as it is insensitive to view-to-view motion. However, the LSDI image is inherently blurred in the line scan direction due to its approximately triangular shape point spread function (PSF). The width of the PSF could be decreased by exciting ultra thin slices. However, to achieve a half maximum width of 1 mm at an LSDI inclination angle of 45, the 90 and 180 slice thickness would have to be 0.7 mm. We investigate the use of a least norm (LN) algorithm to increase the resolution of line scan images without requiring ultra thin slices. Methods and Discussion: Conventional 2DFT image can be represented by a matrix Mi. If the image resolution is high enough compared to the underlying object structure, each element of Mi could be approximated as the amplitude of all the magnetization within that voxel. Under this assumption, when two oblique slices are excited in LSDI and the signal from the cross section is acquired, it could be viewed as several elements in one row of Mi are selected, weighted and summed to form one element in the acquired image Mlsdi. The whole process could be simplified as one matrix operation Mlsdi= A Mi. According to the knowledge of the shape of the PSF, this equation is underdetermined, and there are many solutions for Mi. One of the solutions is the least norm solution Mln =AT(AAT)-1Mlsdi. Mln is a high-resolution image reconstructed from the blurred image Mlsdi.
Showing the most recent 10 out of 446 publications
