Zhaolin Chen

An optimised framework for reconstructing and processing MR phase images.

Phase contrast imaging holds great potential for in vivo biodistribution studies of paramagnetic molecules and materials. However, in vivo quantification of iron storage and other paramagnetic materials requires improvements in reconstruction and processing of MR complex images. To achieve this, we have developed a framework including (i) an optimal coil sensitivity smoothing filter for phase imaging determined at the maximal signal to noise ratio, (ii) a phase optimised and a complex image optimised reconstruction approach, and (iii) a magnitude and phase correlation test criterion to determine the low pass filter parameter for background phase removal. The method has been evaluated using 3T and 7T MRI data containing cortical regions, the basal ganglia including the caudate, and the midbrain including the substantia nigra. The optimised reconstruction improves phase image contrast and noise suppression compared with conventional reconstruction approaches, and the correlation test criterion provides an objective method for separation of the local phase signal from the background phase measurements. Phase values of several brain regions of interest have been calculated, including gray matter (-1.23 Hz at 7T and -0.55 Hz at 3T), caudate (-3.8 Hz at 7T), and the substantia nigra (-6.2 Hz at 7T).

IIR GRAPPA for parallel MR image reconstruction

Accelerated parallel MRI has advantage in imaging speed, and its image quality has been improved continuously in recent years. This paper introduces a two‐dimensional infinite impulse response model of inverse filter to replace the finite impulse response model currently used in generalized autocalibrating partially parallel acquisitions class image reconstruction methods. The infinite impulse response model better characterizes the correlation of k‐space data points and better approximates the perfect inversion of parallel imaging process, resulting in a novel generalized image reconstruction method for accelerated parallel MRI. This k‐space‐based reconstruction method includes the conventional generalized autocalibrating partially parallel acquisitions class methods as special cases and has a new infinite impulse response data estimation mechanism for effective improvement of image quality. The experiments on in vivo MRI data show that the proposed method significantly reduces reconstruction errors compared with the conventional two‐dimensional generalized autocalibrating partially parallel acquisitions method, particularly at the high acceleration rates. Magn Reson Med, 2010.

FB analysis of PMRI and its application to H ∞ optimal sense reconstruction

This paper presents a filter bank (FB) analysis of parallel magnetic resonance imaging (PMRI). The underlying image reconstruction strategies of the most widely used PMRI reconstruction methods are unified within the framework and their fundamental perfect reconstruction (PR) constraints are analyzed. Based on this analysis, an improved reconstruction method, called H∞ optimal sense, is developed and its advantage is demonstrated by an example.

Spatially dependent filtering for removing phase distortions at the cortical surface.

Recent advances in high field magnetic resonance technology have increased the interest in the phase of the complex data. Processed phase images are derived from the phase signal by removing the bias field and phase wraps from the initial data. However, the usefulness of this data has been hindered by artifacts at the brain/non‐brain surface, particularly in cortical regions. A method is proposed that efficiently removes surface artifacts by performing Gaussian filtering with spatially varying parameters of unwrapped or complex filtered phase images. The proposed method is shown to produce improved images, revealing underlying structure and detail that are otherwise obscured by surface artifacts in images produced by traditional phase processing methods. Magn Reson Med, 2011.

PARALLEL MR IMAGE RECONSTRUCTION USING IIR FB

Parallel magnetic resonance imaging (pMRI) has been a very active research area recently due to its advantage of accelerated imaging speed. However, the current image reconstruction methods for pMRI suffer visible artifacts and noise effects. To improve the image quality, this paper proposes a new class of reconstruction methods based on infinite impulse response (IIR) filter bank (FB). The advantages of the proposed methods have been demonstrated by in vivo examples

Adaptive K-space Updating Methods for Dynamic MRI Sequence Estimation

In dynamic magnetic resonance imaging (MRI), due to the limitation of imaging speed, it is commonly required to reconstruct the images from a reduced Fourier encoded K-space sequence. This paper presents two temporal model based methods to estimate the un-acquired K-space data, called adaptive K-space updating (AKU) methods. The AKU reconstruction algorithms are directly applicable to the truncated K-space sequence generated by the well known Fourier keyhole (FK) encoding scheme. The experimental study on the real MRI data shows that the proposed AKU methods can produce images with much lower reconstruction error than conventional FK method

A Method for Shape Analysis and Segmentation in MRI

Morphometry of human magnetic resonance images (MRI) is the process of measuring structural variations that occur in the brain. Morphometrics provide a mechanism to monitor and relate structural changes of anatomy to the onset or progression of a disease. It is therefor a very important area of research, specifically since MRI sequences are non-invasive and can be acquired in-vivo. This paper addresses two sub-problems in the area of MRI morphometry: 1) shape analysis and 2) semi-automated segmentation. Firstly the paper presents a method of analysing for group differences between 2D contours. The theoretical underpinning is derived from the field of content-based image retrieval, specifically to solve contour correspondences. Secondly the paper uses these correspondences to train a deformable model to automatically segment structures. This is achieved using a modified active appearance model fitting algorithm.

Weighted H ∞ optimization approach to parallel MR image reconstruction

Parallel Magnetic Resonance Imaging (pMRI) is one of the most important technical advances in current MRI technology. Although many reconstruction methods have been proposed for pMRI, the reconstructed images still suffer visible artifacts especially when high acceleration is used. To improve the reconstruction quality of pMRI, this paper proposes a new reconstruction method. Instead of achieving perfect reconstruction under ideal assumptions, the proposed method uses weighted H∞ optimization to minimize the total reconstruction error, including that raised by the noise and uncertainty in sensitivity map. The experimental studies on in vivo data sets demonstrate that the proposed method outperforms significantly the existing methods at high acceleration.

2D IIR filter for parallel magnetic resonance image reconstruction

Accelerated parallel MRI has been widely used in medical research and clinical diagnoses. This paper presents a two dimensional (2D) infinite impulse response (IIR) model of inverse filter to replace the finite impulse response model currently used in GRAPPA class image reconstruction methods. The IIR model better characterizes the correlation of k-space data points and better approximates the inversion of parallel imaging process. The experiments on in vivo accelerated cardiac imaging show that the proposed method significantly reduces reconstruction errors compared with conventional 2D GRAPPA method, particularly at the high acceleration rates.

A Multistage Parallel Magnetic Resonance Image Reconstruction Method

The length of scan time is a critical issue in Magnetic Resonance Imaging(MRI). Parallel MRI techniques have recently been introduced to shorten the scanning time by using multiple receiver coils to acquire the MR signal. As a result, the imaging process is accelerated several times compared to conventional non-accelerated MRI. In this paper, we provide a new reconstruction method for parallel MRI that improves the resultant image quality. Our algorithm involves the combination of two popular and commercially utilised parallel MRI reconstruction techniques, SENSE and GRAPPA. These two methods are complementary to each other, and are traditionally implemented in different imaging conditions. Our proposed method, called the SG reconstruction method, takes advantage of the distinct merits of both SENSE and GRAPPA to improve the image reconstruction quality. We demonstrate the superiority of the SG algorithm through comparison to SENSE and GRAPPA applied to high-field, experimental MRI data.