Zen and the art of medical image registration: correspondence, homology, and quality

Crum, William R.; Griffin, Lewis D.; Hill, Derek L. G.; Hawkes, David J.

doi:10.1016/j.neuroimage.2003.07.014

Cited by 160 publications

(116 citation statements)

References 61 publications

Supporting

Mentioning

115

Contrasting

Order By: Relevance

“…Although this technique provides robust normalization of individual subject anatomy to the standard coordinate space, it makes no effort to match brain surface gyri and sulci patterns. Variability of the activation locations in Talairach space due to gyri variability can be reduced by applying more sophisticated normalization techniques with nonlinear warping (Crum et al, 2003). Monitoring of head motion is critically important to ensure robust and reproducible activations.…”

Section: Discussionmentioning

confidence: 99%

A rapid fMRI task battery for mapping of visual, motor, cognitive, and emotional function

2006

View full text Add to dashboard Cite

A set of sensory, motor, cognitive and emotional tasks were combined in a simple, rapid-presentation task battery and tested on a group of 31, normal, healthy subjects aged 22 to 76. Five tasks were selected on the basis of widespread use in fMRI and their ability to produce robust and reliable regional activations. They were: 1) a visual task designed to activate the occipital cortex; 2) a bimanual motor task designed to activate motor areas; 3) a verb generation task designed to activate speech processing areas; 4) an n-back task designed to activate areas associated with working memory and executive function; and 5) an emotional pictures task designed to provoke strong emotional responses that typically activate limbic structures. Most of the tasks produced reliable activations in individual subjects, and assessments of the distribution and reliability of individual subject activations in each targeted area are provided. The emotional pictures task did not demonstrate adequate sensitivity in a priori target regions, only in the a posteriori defined inferior temporal region. Age and gender specific differences were found in the activation patterns for both the cognitive and emotional tasks. The battery provides a prescribed means for researchers to obtain reliable functional localizers within 20-25 minutes of scanning, which can be used to support more elaborate mapping studies of brain function. The dataset can also serve as reliability metric for new fMRI labs and novice investigators seeking to test their acquisition and analysis techniques with minimal time investment and expense.

show abstract

Section: Discussionmentioning

confidence: 99%

A rapid fMRI task battery for mapping of visual, motor, cognitive, and emotional function

2006

View full text Add to dashboard Cite

show abstract

“…Many investigators use SPM2 as a 'black box', influencing its output by the choice of input images and system options, followed by a non-quantitative visual inspection of its results. Some assert that competent identification of errors by visual inspection is a hopeful assumption at best, given the ad hoc nature of these visual approaches (Bookstein, 2001;Crum et al, 2003).…”

Section: Introductionmentioning

confidence: 99%

Statistical parametric mapping of brain morphology: Sensitivity is dramatically increased by using brain-extracted images as inputs

et al. 2006

View full text Add to dashboard Cite

A major attraction of voxel-based morphometry (VBM) is that it allows researchers to explore large datasets with minimal human intervention. However, the validity and sensitivity of the Statistical Parametric Mapping (SPM2) approach to VBM is the subject of considerable debate. We visually inspected the SPM2 gray matter segmentations for 101 research participants and found a gross inclusion of non-brain tissue surrounding the entire brain as gray matter in five subjects, and focal areas bordering the brain in which non-brain tissue was classified as gray matter in many other subjects. We also found many areas in which the cortical grey matter was incorrectly excluded from the segmentation of the brain. The major source of these errors was the misregistration of individual brain images with the reference T1-weighted brain template. These errors could be eliminated if SPM2 operated on images from which non-brain tissues (scalp, skull, and meninges) are removed (brain-extracted images). We developed a modified SPM2 processing pipeline that used brainextracted images as inputs to test this hypothesis. We describe the modifications to the SPM2 pipeline that allow analysis of brain-extracted inputs. Using brain-extracted inputs eliminated the non-brain matter inclusions and the cortical gray matter exclusions noted above, reducing the residual mean square errors (RMSEs, the error term of the SPM2 statistical analyses) by over thirty percent. We show how this reduction in the RMSEs profoundly affects power analyses. SPM2 analyses of brainextracted images may require sample sizes only half as great as analyses of non-brain extracted images.

show abstract

“…One previously proposed method of determining this proximity is the tolerance, τ [6]. The standard overlap measures can be considered to be τ=0 since pairs of label voxels have to occupy the same space to be considered overlapping.…”

Section: Overlap and Target Registration Error: Overlap Tolerancementioning

confidence: 99%

Generalised Overlap Measures for Assessment of Pairwise and Groupwise Image Registration and Segmentation

Crum

Cámara

Rueckert

et al. 2005

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Effective validation techniques are an essential pre-requisite for segmentation and non-rigid registration techniques to enter clinical use. These algorithms can be evaluated by calculating the overlap of corresponding test and gold-standard regions. Common overlap measures compare pairs of binary labels but it is now common for multiple labels to exist and for fractional (partial volume) labels to be used to describe multiple tissue types contributing to a single voxel. Evaluation studies may involve multiple image pairs. In this paper we use results from fuzzy set theory and fuzzy morphology to extend the definitions of existing overlap measures to accommodate multiple fractional labels. Simple formulas are provided which define single figures of merit to quantify the total overlap for ensembles of pairwise or groupwise label comparisons. A quantitative link between overlap and registration error is established by defining the overlap tolerance. Experiments are performed on publicly available labeled brain data to demonstrate the new measures in a comparison of pairwise and groupwise registration.

show abstract

Zen and the art of medical image registration: correspondence, homology, and quality

Cited by 160 publications

References 61 publications

A rapid fMRI task battery for mapping of visual, motor, cognitive, and emotional function

A rapid fMRI task battery for mapping of visual, motor, cognitive, and emotional function

Statistical parametric mapping of brain morphology: Sensitivity is dramatically increased by using brain-extracted images as inputs

Generalised Overlap Measures for Assessment of Pairwise and Groupwise Image Registration and Segmentation

Contact Info

Product

Resources

About