Limits...
Subject-Motion Correction in HARDI Acquisitions: Choices and Consequences.

Elhabian S, Gur Y, Vachet C, Piven J, Styner M, Leppert IR, Pike GB, Gerig G - Front Neurol (2014)

Bottom Line: The scientific community offers strong theoretical and experimental work on noise reduction and orientation distribution function (ODF) reconstruction techniques for HARDI data, where post-acquisition motion correction is widely performed, e.g., using the open-source DTIprep software (1), FSL (the FMRIB Software Library) (2), or TORTOISE (3).Nonetheless, effects and consequences of the selection of motion correction schemes on the final analysis, and the eventual risk of introducing confounding factors when comparing populations, are much less known and far beyond simple intuitive guessing.Effects of various motion correction choices are systematically explored and illustrated, leading to a general conclusion of discouraging users from setting ad hoc thresholds on the estimated motion parameters beyond which volumes are claimed to be corrupted.

View Article: PubMed Central - PubMed

Affiliation: Scientific Computing and Imaging Institute , Salt Lake City, UT , USA ; Faculty of Computers and Information, Cairo University , Cairo , Egypt.

ABSTRACT
Diffusion-weighted imaging (DWI) is known to be prone to artifacts related to motion originating from subject movement, cardiac pulsation, and breathing, but also to mechanical issues such as table vibrations. Given the necessity for rigorous quality control and motion correction, users are often left to use simple heuristics to select correction schemes, which involves simple qualitative viewing of the set of DWI data, or the selection of transformation parameter thresholds for detection of motion outliers. The scientific community offers strong theoretical and experimental work on noise reduction and orientation distribution function (ODF) reconstruction techniques for HARDI data, where post-acquisition motion correction is widely performed, e.g., using the open-source DTIprep software (1), FSL (the FMRIB Software Library) (2), or TORTOISE (3). Nonetheless, effects and consequences of the selection of motion correction schemes on the final analysis, and the eventual risk of introducing confounding factors when comparing populations, are much less known and far beyond simple intuitive guessing. Hence, standard users lack clear guidelines and recommendations in practical settings. This paper reports a comprehensive evaluation framework to systematically assess the outcome of different motion correction choices commonly used by the scientific community on different DWI-derived measures. We make use of human brain HARDI data from a well-controlled motion experiment to simulate various degrees of motion corruption and noise contamination. Choices for correction include exclusion/scrubbing or registration of motion corrupted directions with different choices of interpolation, as well as the option of interpolation of all directions. The comparative evaluation is based on a study of the impact of motion correction using four metrics that quantify (1) similarity of fiber orientation distribution functions (fODFs), (2) deviation of local fiber orientations, (3) global brain connectivity via graph diffusion distance (GDD), and (4) the reproducibility of prominent and anatomically defined fiber tracts. Effects of various motion correction choices are systematically explored and illustrated, leading to a general conclusion of discouraging users from setting ad hoc thresholds on the estimated motion parameters beyond which volumes are claimed to be corrupted.

No MeSH data available.


Related in: MedlinePlus

Average and standard deviation of the percentage of motion-corrupted gradient directions as a function of thresholding on the estimated rotation angle in degrees (left) and the estimated translation magnitude in millimeter (right) for three human phantoms scanned twice at four clinical sites. The boxplots show the overall statistics of estimated motion parameters.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4260507&req=5

Figure 2: Average and standard deviation of the percentage of motion-corrupted gradient directions as a function of thresholding on the estimated rotation angle in degrees (left) and the estimated translation magnitude in millimeter (right) for three human phantoms scanned twice at four clinical sites. The boxplots show the overall statistics of estimated motion parameters.

Mentions: To quantify motion, we used the magnitude of the translation vector (in millimeters) as well as the axis–angle rotation representation (in degrees) (4). The boxplots in Figure 2 show the rotational and translational components of the motion being detected from a total of 24 DWI datasets, showing an average of 0.39° rotation and 0.61 mm translation. The graphs in Figure 2 illustrate the arbitrariness of a common calculation of percentage of motion correction to determine the number of affected scans, here shown as a function of thresholding on the estimated motion parameters. While this experiment attributes the estimated rotation and translation parameters to actual subject motion, a part of the experimentally obtained parameters may be due to some imaging/image-processing uncertainty and also to image differences due to anatomical properties of the object (e.g., tissue orientation) that make the images “look” different even if they were perfectly aligned. To backup our analysis, we conducted another experiment where we contaminated a single DWI dataset with two independent realizations of Rician noise such that the two generated DWI images were perfectly aligned because they were the exact same image. Then, we ran motion correction where all DWI images were aligned to the same baseline, we obtain similar motion parameters although we are registering two independent acquisitions of the same subject. We therefore conclude that the transformation parameter estimates from FSL-MCFLIRT (31) are resilient to noise and may primarily caused by subject motion during a DWI scan, or eventually also by relative motion between subject and scans if considering artifacts due to pulse sequence and scanner technology.


Subject-Motion Correction in HARDI Acquisitions: Choices and Consequences.

Elhabian S, Gur Y, Vachet C, Piven J, Styner M, Leppert IR, Pike GB, Gerig G - Front Neurol (2014)

Average and standard deviation of the percentage of motion-corrupted gradient directions as a function of thresholding on the estimated rotation angle in degrees (left) and the estimated translation magnitude in millimeter (right) for three human phantoms scanned twice at four clinical sites. The boxplots show the overall statistics of estimated motion parameters.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4260507&req=5

Figure 2: Average and standard deviation of the percentage of motion-corrupted gradient directions as a function of thresholding on the estimated rotation angle in degrees (left) and the estimated translation magnitude in millimeter (right) for three human phantoms scanned twice at four clinical sites. The boxplots show the overall statistics of estimated motion parameters.
Mentions: To quantify motion, we used the magnitude of the translation vector (in millimeters) as well as the axis–angle rotation representation (in degrees) (4). The boxplots in Figure 2 show the rotational and translational components of the motion being detected from a total of 24 DWI datasets, showing an average of 0.39° rotation and 0.61 mm translation. The graphs in Figure 2 illustrate the arbitrariness of a common calculation of percentage of motion correction to determine the number of affected scans, here shown as a function of thresholding on the estimated motion parameters. While this experiment attributes the estimated rotation and translation parameters to actual subject motion, a part of the experimentally obtained parameters may be due to some imaging/image-processing uncertainty and also to image differences due to anatomical properties of the object (e.g., tissue orientation) that make the images “look” different even if they were perfectly aligned. To backup our analysis, we conducted another experiment where we contaminated a single DWI dataset with two independent realizations of Rician noise such that the two generated DWI images were perfectly aligned because they were the exact same image. Then, we ran motion correction where all DWI images were aligned to the same baseline, we obtain similar motion parameters although we are registering two independent acquisitions of the same subject. We therefore conclude that the transformation parameter estimates from FSL-MCFLIRT (31) are resilient to noise and may primarily caused by subject motion during a DWI scan, or eventually also by relative motion between subject and scans if considering artifacts due to pulse sequence and scanner technology.

Bottom Line: The scientific community offers strong theoretical and experimental work on noise reduction and orientation distribution function (ODF) reconstruction techniques for HARDI data, where post-acquisition motion correction is widely performed, e.g., using the open-source DTIprep software (1), FSL (the FMRIB Software Library) (2), or TORTOISE (3).Nonetheless, effects and consequences of the selection of motion correction schemes on the final analysis, and the eventual risk of introducing confounding factors when comparing populations, are much less known and far beyond simple intuitive guessing.Effects of various motion correction choices are systematically explored and illustrated, leading to a general conclusion of discouraging users from setting ad hoc thresholds on the estimated motion parameters beyond which volumes are claimed to be corrupted.

View Article: PubMed Central - PubMed

Affiliation: Scientific Computing and Imaging Institute , Salt Lake City, UT , USA ; Faculty of Computers and Information, Cairo University , Cairo , Egypt.

ABSTRACT
Diffusion-weighted imaging (DWI) is known to be prone to artifacts related to motion originating from subject movement, cardiac pulsation, and breathing, but also to mechanical issues such as table vibrations. Given the necessity for rigorous quality control and motion correction, users are often left to use simple heuristics to select correction schemes, which involves simple qualitative viewing of the set of DWI data, or the selection of transformation parameter thresholds for detection of motion outliers. The scientific community offers strong theoretical and experimental work on noise reduction and orientation distribution function (ODF) reconstruction techniques for HARDI data, where post-acquisition motion correction is widely performed, e.g., using the open-source DTIprep software (1), FSL (the FMRIB Software Library) (2), or TORTOISE (3). Nonetheless, effects and consequences of the selection of motion correction schemes on the final analysis, and the eventual risk of introducing confounding factors when comparing populations, are much less known and far beyond simple intuitive guessing. Hence, standard users lack clear guidelines and recommendations in practical settings. This paper reports a comprehensive evaluation framework to systematically assess the outcome of different motion correction choices commonly used by the scientific community on different DWI-derived measures. We make use of human brain HARDI data from a well-controlled motion experiment to simulate various degrees of motion corruption and noise contamination. Choices for correction include exclusion/scrubbing or registration of motion corrupted directions with different choices of interpolation, as well as the option of interpolation of all directions. The comparative evaluation is based on a study of the impact of motion correction using four metrics that quantify (1) similarity of fiber orientation distribution functions (fODFs), (2) deviation of local fiber orientations, (3) global brain connectivity via graph diffusion distance (GDD), and (4) the reproducibility of prominent and anatomically defined fiber tracts. Effects of various motion correction choices are systematically explored and illustrated, leading to a general conclusion of discouraging users from setting ad hoc thresholds on the estimated motion parameters beyond which volumes are claimed to be corrupted.

No MeSH data available.


Related in: MedlinePlus