Maximum-likelihood estimation of Rician distribution parameters

Sijbers, J.; Dekker, Arnold J. den; Scheunders, Paul; Dyck, D. Van

doi:10.1109/42.712125

Cited by 348 publications

(267 citation statements)

References 16 publications

Supporting

Mentioning

254

Contrasting

Unclassified

Order By: Relevance

“…In this study, real images were obtained by FSE-IR acquisition for T 1 measurement; however, only magnitude images described by the Rician distribution could be obtained by MSE acquisition for T 2 measurement because of restrictions of the MRI apparatus. Because magnitude images described by the Rician distribution with insufficient SNR may contribute to inaccuracy of T 2 measurement (35,36), special attention should be paid to maintain a sufficient SNR when multislice acquisition is used for T 2 measurement.…”

Section: Discussionmentioning

confidence: 99%

“…A total of 10 knees were involved in the T 1 measurement study (three men, two women), and the other 10 knees were involved in the T 2 measurement study (three men, two women). Mean age at time of imaging was 31.8 Ϯ 3.9 years [25][26][27][28][29][30][31][32][33][34][35][36] (values are given throughout as mean Ϯ SD [range]) for the T 1 measurement study and 29.5 Ϯ 4.9 years [24 -36] for the T 2 measurement study. Exclusion criteria were history of knee pain or abnormality or trauma of the knee joint requiring medical treatment.…”

Section: Volunteersmentioning

confidence: 99%

See 1 more Smart Citation

Effect of multislice acquisition on T₁ and T₂ measurements of articular cartilage at 3T

Watanabe

Boesch

Obata³

et al. 2007

Magnetic Resonance Imaging

View full text Add to dashboard Cite

Purpose: The aim of this study was to investigate the effect of magnetization transfer on multislice T 1 and T 2 measurements of articular cartilage. Materials and Methods:A set of phantoms with different concentrations of collagen and contrast agent (Gd-DTPA 2-) were used for the in vitro study. A total of 20 healthy knees were used for the in vivo study. T 1 and T 2 measurements were performed using fast-spin-echo inversion-recovery (FSE-IR) sequence and multi-spin-echo (MSE) sequence, respectively, in both in vitro and in vivo studies. We investigated the difference in T 1 and T 2 values between that measured by singleslice acquisition and that measured by multislice acquisition.Results: Regarding T 1 measurement, a large drop of T 1 in all slices and also a large interslice variation in T 1 were observed when multislice acquisition was used. Regarding T 2 measurement, a substantial drop of T 2 in all slices was observed; however, there was no apparent interslice variation when multislice acquisition was used. Conclusion:This study demonstrated that the adaptation of multislice acquisition technique for T 1 measurement using FSE-IR methodology is difficult and its use for clinical evaluation is problematic. In contrast, multislice acquisition for T 2 measurement using MSE was clinically applicable if inaccuracies caused by multislice acquisition were taken into account. IN RECENT YEARS, several qualitative MRI techniques have been developed to monitor the composition and structure of articular cartilage (1,2). Because early degeneration of cartilage is characterized by deterioration of the extracellular matrix components (3-5) such as glycosaminoglycan (GAG) and collagen, qualitative MRI techniques have the potential to identify cartilage degeneration in an early stage. Delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) is a qualitative MRI technique that can evaluate the concentration of GAG in cartilage (6,7). Transverse relaxation time (T 2 ) mapping is an MRI technique that can evaluate the integrity of the collagen network structure and water content in cartilage (8,9). These two MRI techniques have had their capacities validated in basic studies (6,8,10,11) and have been used for clinical evaluations (12-15). The dGEMRIC technique requires longitudinal relaxation time (T 1 ) measurement, whereas T 2 mapping requires T 2 measurement; both techniques employ dedicated pulse sequences. There are several kinds of pulse sequences for T 1 or T 2 measurements. Fast-spin-echo inversion-recovery (FSE-IR) sequence and multi-spinecho (MSE) sequence have been commonly used for T 1 measurement and T 2 measurement of articular cartilage, respectively (7,9,(12)(13)(14)(15). The main reason to use these sequences for clinical evaluation is that they offer acceptable calculation accuracy and reproducibility in a suitable acquisition time. However, it is known that several undesirable factors, which can cause inaccuracy in T 1 and T 2 measurements, may be introduced when these sequences are used for multislice rather ...

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Volunteersmentioning

confidence: 99%

Effect of multislice acquisition on T₁ and T₂ measurements of articular cartilage at 3T

Watanabe

Boesch

Obata³

et al. 2007

Magnetic Resonance Imaging

View full text Add to dashboard Cite

show abstract

“…In non-zero signal regions, signals are Rician distributed (12,13). In the signal regions where SNR Ͼ 5, the noise is again approximately Gaussian (8,9).…”

Section: Signal Behavior In Subtractionsmentioning

confidence: 99%

Analysis of subtraction methods in three‐dimensional contrast‐enhanced peripheral MR angiography

Huang

Webster

Wright

2002

Magnetic Resonance Imaging

View full text Add to dashboard Cite

Purpose:To compare the effectiveness of three image subtraction algorithms designed to improve arterial conspicuity in first-pass contrast-enhanced magnetic resonance (MR) angiography. Materials and Methods:Three subtraction methods were analyzed through computer simulations, phantom studies, and clinical studies. These algorithms were: complex subtraction, magnitude subtraction, and maximum intensity projection subtraction. Results:In high resolution three-dimensional imaging, maximum intensity projection subtraction generally yields the best background suppression. Complex subtraction is effective in reducing partial volume effects in low resolution imaging. Magnitude subtraction works better in high resolution, low contrast concentration protocols. Conclusion:Choosing the appropriate subtraction method according to the protocol is helpful in optimizing image quality. SUBTRACTION IS A COMMONLY used technique to increase arterial conspicuity in contrast-enhanced magnetic resonance (MR) angiography. Although the importance of complex subtraction has been identified (1), magnitude images reconstructed by the scanner were generally used for the subtraction in most published papers (2-5). In theory, subtracting complex numbers can reduce partial volume effects and recover some vessel signals. However, we found complex subtraction did not always give us the best result in threedimensional MR angiography. On the contrary, subtracting two maximum intensity projection (MIP) images, which was thought to be "not useful" (6), often yielded the best arterial conspicuity in a multiple injection protocol (Fig. 1). This suggests that different subtraction algorithms should be applied in different situations. An analysis of the underlying reason for different results through mathematical modeling and clinical data evaluation is needed. This study will provide a theoretical basis for choosing the appropriate subtraction algorithm for MR angiography. Key THEORY SubtractionTwo characteristics of the MR data sets contribute to the multiplicity of choices for subtraction algorithms. First, MR raw data are complex numbers. We can therefore either subtract the complex numbers of the two corresponding data sets pixel by pixel and then take the magnitude, or we can first take the magnitude of the two data sets and then subtract these magnitude values. In MR imaging, the difference of order could be significant in the presence of partial volume effects (1).The second characteristic is that we use a threedimensional acquisition to cover the vasculature. Thus, some kind of volume rendering algorithm is needed to convert the three-dimensional data set into two-dimensional form to be displayed on a computer screen or film. MIP is the most popular volume rendering algorithm in three-dimensional MR angiography because it is operator independent, free of intensity thresholding, and relatively fast. Although MIP has some potential problems, no other algorithm is currently more practical and objective in the overall sense. We can either first do...

show abstract

“…Noise in MRI magnitude equation is also very similar to equation (2); where noise is both signal dependent and additive. At the same time Rician noise is also nonlinear due to magnitude image formation, from information of real and imaginary components of the signal with respect to noise [8].…”

Section: Signal-dependent Filter and Noisementioning

confidence: 99%

“…The filtered cartilage image has been compared with the observed image, due to absence of true noise free image of the knee. The noise estimated in the knee MRI from the background is of standard deviation 13.54 and that for the spine image is 6.69 using equation (8). A plot of PSNR versus standard deviation of noise was also prepared for study of the denoising process at higher values of noise, using the spine image.…”

Section: Illustrative Simulationmentioning

confidence: 99%

Signal Dependent Rician Noise Denoising Using Nonlinear Filter

Aarya¹,

Jiang²,

Gale³

2013

LNSE

View full text Add to dashboard Cite

Abstract-MR images are increasingly used for diagnostic and surgical procedures, as they offer better soft tissue contrast and advanced imaging capabilities. Similar to other imaging modalities, MR images are also subjected to various forms of noises and artifacts. The noise affecting MRI images is known as Rician noise and displays a nonlinear and signal dependent behavior. In this paper we propose a nonlinear filtering method for Rician noise denoising. Nonlinear filters are more capable in addressing signal dependent behavior of noise and offer good denoising with better edge preserving capabilities. A nonlinear filter based on homomorphic filter characteristics has been designed to address Rician noise in MR images. The proposed filter has been implemented on synthetic images and MR images of the articular cartilage. The efficiency of the proposed filtering method is verified by computing the PSNR and SSIM index of the image. The proposed nonlinear filter performs good denoising with improvement in the image quality as observed from the PSNR values of the image. It also offers edge preservation and can be used for both structural MRI and soft tissue study effectively Index Terms-Homomorphic filters, MRI denoising, Rician noise, signal dependent filtering.

show abstract

Maximum-likelihood estimation of Rician distribution parameters

Cited by 348 publications

References 16 publications

Effect of multislice acquisition on T₁ and T₂ measurements of articular cartilage at 3T

Effect of multislice acquisition on T₁ and T₂ measurements of articular cartilage at 3T

Analysis of subtraction methods in three‐dimensional contrast‐enhanced peripheral MR angiography

Signal Dependent Rician Noise Denoising Using Nonlinear Filter

Contact Info

Product

Resources

About

Maximum-likelihood estimation of Rician distribution parameters

Cited by 348 publications

References 16 publications

Effect of multislice acquisition on T1 and T2 measurements of articular cartilage at 3T

Effect of multislice acquisition on T1 and T2 measurements of articular cartilage at 3T

Analysis of subtraction methods in three‐dimensional contrast‐enhanced peripheral MR angiography

Signal Dependent Rician Noise Denoising Using Nonlinear Filter

Contact Info

Product

Resources

About

Effect of multislice acquisition on T₁ and T₂ measurements of articular cartilage at 3T

Effect of multislice acquisition on T₁ and T₂ measurements of articular cartilage at 3T