“Squashing peanuts and smashing pumpkins”: How noise distorts diffusion‐weighted MR data

By Derek K. Jones, Peter J. Basser. Magnetic Resonance in Medicine, 2004

630 citations recorded by the source corpus.

New diffusion‐weighted imaging (DWI) methods, including high‐b, q‐space, and high angular resolution MRI methods, attempt to extract information about non‐Gaussian diffusion in tissue that is not provided by low‐b‐value (b ≈ 1000 s mm−2) diffusion or diffusion tensor magnetic resonance imaging (DT‐MRI). Additionally, DWI data with higher spatial resolution are being acquired to resolve fine anatomic structures, such as white matter fasciculi. Increasing diffusion‐weighting or decreasing voxel size can reduce the signal‐to‐noise ratio so that some DWI signals are close to the background noise level. Here we report several new artifacts that can be explained by considering how background noise affects the peanut‐shaped angular apparent diffusion coefficient (ADC) profile. These include an orientationally dependent deviation from Gaussian behavior of the ADC profile, an underestimation of indices of diffusion anisotropy, and a correlation between estimates of mean diffusivity and diffusion anisotropy. We also discuss how noise can cause increased gray/white matter DWI contrast at higher b values and an apparent elevation of diffusion anisotropy in acute ischemia. Importantly, all of these artifacts are negligible in the b‐value range typically used in DT‐MRI of brain (b ≈ 1000 s mm−2). Finally, we demonstrate a strategy for ameliorating the rectified noise artifact in data collected at higher b values. Magn Reson Med 52:979–993, 2004. Published 2004 Wiley‐Liss, Inc.

Subfields: Orthopedics and Sports Medicine, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/bf1f24263832cdeeedbcad36ad9ed45ba306995f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Orthopedics and Sports Medicine

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Compartment models of the diffusion MR signal in brain white matter: A taxonomy and comparison (2012, neutral): “, 1988) for minimising a chi-squared objective function with the offset Gaussian noise model, which adds an expected offset to each measurement from Rician noise bias (Alexander, 2009; Jones and Basser, 2004).”
Diffusion Weighted Image Denoising Using Overcomplete Local PCA (2013, background): “…over small local windows instead of the whole image what significantly produces sparser representation (in the extreme case, i.e. in homogeneous areas, it can result in the sparsest representation not requiring any component to represent the data but on only the mean value of the region).”
NODDI: Practical in vivo neurite orientation dispersion and density imaging of the human brain (2012, background): “The dependence of the bias on the groundtruth value of OD mirrors the noise-induced bias in the estimate of FA (Jones and Basser, 2004; Pierpaoli and Basser, 1996): for low FA, corresponding to high OD, the bias is negative and its magnitude increases as the ground-truth value of FA decreases.”
Weighted linear least squares estimation of diffusion MRI parameters: Strengths, limitations, and pitfalls (2013, background): “The SNR is here defined as the ratio between the non-DW signal and the noise level (cf. Jones and Basser, 2004).”
Ultra-Fast MRI of the Human Brain with Simultaneous Multi-Slice Imaging (2013, background): “This significantly larger SNR per shot results in less sensitivity to the non-Gaussian noise behavior and the resulting bias in the estimation of diffusion parameters [10].”
Denoising of diffusion MRI using random matrix theory (2016, background): “Indeed, many noise-related artifacts have been discussed earlier (Jones and Basser, 2004; Veraart et al., 2013a, 2011a,b).”
Artifacts and pitfalls in diffusion MRI (2006, background): “Furthermore, although anisotropy may be small in voxels that contain merging or splitting fibers, the tracking process should not be stopped at such locations (68).”
Model-based analysis of multi-shell diffusion MR data for tractography: How to get over fitting problems (2012, background): “This is similar in spirit to the maximum-likelihood approach introduced for the diffusion tensor model in [20].”
Anatomically-constrained tractography: Improved diffusion MRI streamlines tractography through effective use of anatomical information (2012, methodology): “The diffusion tensor was the first proposed model for characterising the underlying tissuemicrostructure in terms of themeasured diffusion tography; b-value, amount of with zero diffusion-weighting; onally-Encoded Colour; dFOD, Distributions; DW, diffusionplanar imaging; FA, Fractional cking; FOD, F”
Pushing the limits of in vivo diffusion MRI for the Human Connectome Project (2013, methodology): “(Jones and Basser, 2004; Koay et al., 2009) The use of a 64 channel array can increase the degree of magnitude bias present and care was taken to use a complex weighted coil combination instead of a simple “sum of squares” combination (Sotiropoulos et al., 2013).”
White matter integrity, fiber count, and other fallacies: The do's and don'ts of diffusion MRI (2013, background): “diffusion-weighted images — so as to avoid the problems associated with the rectifiednoise floor (Jones and Basser, 2004) and, as stated ear-”
A modified damped Richardson-Lucy algorithm to reduce isotropic background effects in spherical deconvolution (2010, background): “Due to the Rician distribution of the MRI noise, a minimum signal level is always present in diffusion-weighted images (Clarke et al., 2008; Jones and Basser, 2004).”
Partial volume effect as a hidden covariate in DTI analyses (2011, background): “0032 mm(2)/s); and one representing the cingulum environment (in the following referred to as “cingulum simulations”, with background FA and MD values equal to the values within the fiber bundle, albeit with a different orientation) (Le Bihan et al., 2001; Jones and Basser, 2004).”
Improving diffusion MRI using simultaneous multi-slice echo planar imaging (2012, background): “When diffusion weighted image (DWI) signals are close to the back-ground noise level, deviation from Gaussian noise behavior can cause signal bias that is orientationally dependent, resulting in a biased estimate of the diffusion parameters (Jones and Basser, 2004).”
Effects of diffusion weighting schemes on the reproducibility of DTI-derived fractional anisotropy, mean diffusivity, and principal eigenvector measurements at 1.5T (2007, methodology): “) which result from the tensor formalism (Batchelor et al., 2003; Hasan et al., 2004; Jones, 2003; Jones and Basser, 2004).”
Imaging brain microstructure with diffusion MRI: practicality and applications (2019, background): “[230]), the range in which parameters can be estimated with high accuracy [231, 232], and the interplay between the hardware constraints and parameter estimates, e.”
Characterization of Cerebral White Matter Properties Using Quantitative MRI Stains (2011, methodology): “For DTI studies, the diffusion tensor is often estimated using a linear least squares or weighted least squares approach, though a nonlinear least squares approach has been shown to yield the most robust fit in noisy data ( Jones and Basser, 2004; Koay et al., 2006).”