SHORE‐based detection and imputation of dropout in diffusion MRI

Abstract

Purpose In diffusion MRI, dropout refers to a strong attenuation of the measured signal that is caused by bulk motion during the diffusion encoding. When left uncorrected, dropout will be erroneously interpreted as high diffusivity in the affected direction. We present a method to automatically detect dropout, and to replace the affected measurements with imputed values. Methods Signal dropout is detected by deriving an outlier score from a simple harmonic oscillator‐based reconstruction and estimation (SHORE) fit of all measurements. The outlier score is defined to detect measurements that are substantially lower than predicted by SHORE in a relative sense, while being less sensitive to measurement noise in cases of weak baseline signal. A second SHORE fit is based on detected inliers only, and its predictions are used to replace outliers. Results Our method is shown to reliably detect and accurately impute dropout in simulated data, and to achieve plausible results in corrupted in vivo dMRI measurements. Computational effort is much lower than with previously proposed alternatives. Conclusions Deriving a suitable outlier score from SHORE results in a fast and accurate method for detection and imputation of dropout in diffusion MRI. It requires measurements with multiple b values (such as multi‐shell or DSI), but is independent from the models used for analysis (such as DKI, NODDI, deconvolution, etc.).

Incorporating outlier information into diffusion MR tractogram filtering for robust structural brain connectivity and microstructural analyses

“…It can be reasonably argued that the rather naive outlier replacement we implemented here could be improved with already available proposals of q-space neighborhood (Niethammer et al, 2007) or model based estimations (Lauzon et al, 2013;Andersson et al, 2016;Koch et al, 2019). However, same applies to the in-house robust spherical harmonic linear least squares estimator we developed for this task which simply down weighs the outlier measurements.…”

“…the mean or median value of the sample) to complex model prediction based replacements. Some of these ideas have already been transferred to dMRI usage by replacing outliers by their q-space neighborhood (Niethammer et al, 2007) or model based estimations (Lauzon et al, 2013;Andersson et al, 2016;Koch et al, 2019). These methods have in common that they depend on perfect outlier detection which can be problematic if there are many outliers.…”

What’s new and what’s next in diffusion MRI preprocessing

“…TORTOISE employs the MAPRMI model ( Özarslan et al, 2013 ) to predict the diffusion signal for motion and eddy currents distortion correction but does not replace the signal with the predicted values, instead, relies on RESTORE and iRESTORE ( Chang et al, 2005 ; 2012 ) robust fitting approaches. Koch et al (2019) derive an outlier score from a harmonic oscillator-based reconstruction and estimation (SHORE) fit of all measurements ( Özarslan et al, 2013 ), and achieve a higher accuracy compared to Gaussian Process-based outlier detection with lower computational demands. Christiaens et al (2021) integrate slice-wise outlier detection with within-volume motion correction and use their SHARD signal representation to obtain predictions.…”

Fast qualitY conTrol meThod foR derIved diffUsion Metrics (YTTRIUM) in big data analysis: U.K. Biobank 18,608 example

“…In turn, optimised post‐processing pipelines (Ades‐Aron et al, 2018; Maximov, Alnæs, & Westlye, 2019; Tournier et al 2019) and stringent procedures for quality control (QC; Alfaro‐Almagro et al, 2018; Bastiani et al, 2019; Graham, Drobnjak, & Zhang, 2018; Haddad et al, 2019) are important to increase reliability and sensitivity. Various approaches have been developed to detect and correct artefacts in raw diffusion data originating, for example, from eddy currents, bulk head motions, susceptibility distortions (Andersson & Sotiropoulos, 2016), noise (Kochunov et al, 2018), Gibbs ringing artefacts (Perrone et al, 2016; Veraart, Fieremans, Jelescu, Knoll, & Novikov, 2016; Veraart, Novikov, et al, 2016), presence of outliers (Koch, Zhukov, Stöcker, Groeschel, & Schultz, 2019) and diffusion metric variability (David, Mesri, Viergever, & Leemans, 2019; Maximov et al, 2015).…”