Avoiding asymmetry-induced bias in longitudinal image processing

By Martin Reuter, Bruce Fischl. NeuroImage, 2011

503 citations recorded by the source corpus.

Longitudinal image processing procedures frequently transfer or pool information across time within subject, with the dual goals of reducing the variability and increasing the accuracy of the derived measures. In this note, we discuss common difficulties in longitudinal image processing, focusing on the introduction of bias, and describe the approaches we have taken to avoid them in the FreeSurfer longitudinal processing stream.

Subfields: Computer Vision and Pattern Recognition, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/ce1d5efcd691b06d245a3edbeafc104317068733

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Statistics and Probability

Citations in the corpus

Brain templates and atlases (2012, background): “…ethnicity and disease (Geschwind and Galaburda, 1985; Good et al., 2001; Im et al., 2010; LeMaitre et al., 2005; Luders and Toga, 2010; Luders et al., 2004; Reuter and Fischl, 2011; Thiebaut de Schotten et al., 2011; Thompson et al., 1998; Toga and Thompson, 2003; Zilles et al., 1997, 2001).”
Within-subject template estimation for unbiased longitudinal image analysis (2012, background): “In Reuter and Fischl (2011) we argue that avoiding only resampling bias may not be sufficient as different sources of bias exist, such as informing processes in follow-up images with information from baseline.”
Longitudinal multiple sclerosis lesion segmentation: Resource and challenge (2017, background): “Our registration steps are inverse consistent and thus any registration based biases are avoided (Reuter and Fischl, 2011).”
Unique developmental trajectories of cortical thickness and surface area (2014, unknown): “In order to reduce within subject scan session variability, a longitudinal stream was developed for FreeSurfer by Reuter and Fischl (2011).”
Growth of language-related brain areas after foreign language learning (2012, background): “The longitudinal stream creates an unbiased (from the two timepoints) within-subject image (Reuter and Fischl, 2011), increasing reliability and statistical power compared to the regular FreeSurfer stream.”
Symmetric Diffeomorphic Modeling of Longitudinal Structural MRI (2013, background): “The increasingly signiﬁcant potential of longitudinal MRI for clinical diagnosis, as a biomarker of disease progression and as an outcome measure in treatment trials, has led to an increased focus on this issue (Fox et al., 2011; Reuter and Ashburner and Ridgway”
Statistical Analysis of Longitudinal Neuroimage Data with Linear Mixed Effects Models (2012, methodology): “edu/fswiki/LongitudinalProcessing) (Reuter and Fischl, 2011; Reuter et al., 2010; Reuter et al., 2012).”
Models of the Aging Brain Structure and Individual Decline (2012, background): “Thus it needs highly sophisticated preprocessing pipelines in order to detect the changes of interest and achieve unbiased results (Reuter et al., 2010; Reuter and Fischl, 2011).”
FreeSurfer (2012, background): “, 2009), and longitudinal analysis (Reuter and Fischl, 2011; Reuter et al., 2010).”
Image processing and analysis methods for the Adolescent Brain Cognitive Development Study (2018, background): “13 variability and increases sensitivity in the measurement of changes in cortical thickness and the volumes of subcortical structures (Reuter and Fischl, 2011; Reuter et al., 2010, 2012).”
CAT – A Computational Anatomy Toolbox for the Analysis of Structural MRI Data (2023, unknown): “In practice, however, longitudinal data strongly benefit from workflows specifically tailored towards longitudinal analyses, where MR-based noise and inhomogeneities are further reduced and where spatial correspondences are ensured, the latter not only across subjects but also across time points wit”