Euromedim 2006 :1st European Conference on Molecular Imaging Technology

MAGNETIC RESONANCE IMAGE RECONSTRUCTION USING ANALYTIC IMAGE REPRESENTATION

10 May 2006, 17:30

15m

Palais du Pharo, Marseille

oral S3_S4 others modalities Other Modalities

Speaker

Ms Josiane YANKAM NJIWA (CREATIS)

Description

Partial k-space acquisition is a subject of great interest in MRI. This type of acquisition usually consists of, in addition to the positive half of k -space, acquiring a number of negative frequencies beyond the k-space center, thus requiring an increase of acquisition time. Furthermore, the quality of the reconstructed image depends on the number of negative frequencies acquired. To date it has not been demonstrated that the non acquisition of negative frequencies of k-space can produce useful reconstructed images. We propose a new approach for reconstructing MR images from partial k-space. It is based on the notion of analytic image and presents the particularity of using only exactly half of k-space without involving any negative frequencies. We first give a systematic formulation of the notion of analytic image, by defining that an analytic image associated to a real image is a complex image whose real part is equal to the real image to be reconstructed, and whose imaginary part is the two- dimensional Hilbert transform of the real part. Equivalently, in k-space, the Fourier transform (FT) of an analytic image is zero for the negative (or positive) half of k-space, and twice the FT of the analytic image’s real part for the positive frequencies. So, in practice, given a half of k-space, we first consider the omitted half k-space as zero, and multiply the acquired k-space samples by a factor of 2. Then, the thus obtained k-space is inverse Fourier transformed, and the image is reconstructed by taking the magnitude of the complex image. The proposed MR reconstruction method was evaluated on both simulated and real MR data. For simulated data, a slowly varying phase was introduced to generate a complex image whose k-space does not exhibit Hermitian symmetry. For the real MR data, the raw k-space data were acquired using a GE 1.5 T MRI system. The reconstruction quality was assessed both visually and using quantitative criteria such as root mean square (RMS). The obtained results showed that, with the conventional conjugate symmetry method (without using negative frequencies as well as phase estimation), the reconstruction quality depends on the type of images and characteristics of the undesired phase shifting, and is characterized by the presence of important artefacts as reported in the literature, whereas with the proposed method, the reconstruction quality is fairly good and stable. Moreover, in all cases, the proposed method exhibited a substantially better reconstruction quality than the conjugate symmetry method. So, the proposed method based on analytic image representation offers the possibility of reconstructing acceptable images using only one and exactly one half of k-space without involving any negative frequencies, or phase correction.