Limits...
Reliable measurements of brain atrophy in individual patients with multiple sclerosis

View Article: PubMed Central - PubMed

ABSTRACT

Introduction: As neurodegeneration is recognized as a major contributor to disability in multiple sclerosis (MS), brain atrophy quantification could have a high added value in clinical practice to assess treatment efficacy and disease progression, provided that it has a sufficiently low measurement error to draw meaningful conclusions for an individual patient.

Metrixmetrix: In this paper, we present an automated longitudinal method based on Jacobian integration for measuring whole‐brain and gray matter atrophy based on anatomical magnetic resonance images (MRI), named MS. MS is specifically designed to measure atrophy in patients with MS, by including iterative lesion segmentation and lesion filling based on FLAIR and T1‐weighted MRI scans.

Metrixmetrixmetrixmetrix: MS is compared with SIENA with respect to test–retest error and consistency, resulting in an average test–retest error on an MS data set of 0.13% (MS) and 0.17% (SIENA) and a consistency error of 0.07% (MS) and 0.05% (SIENA). On a healthy subject data set including physiological variability the test–retest is 0.19% (MS) and 0.31% (SIENA).

Metrix: Therefore, we can conclude that MS could be of added value in clinical practice for the follow‐up of treatment and disease progression in MS patients.

No MeSH data available.


First row: Comparison of the measurement error of the longitudinal and cross‐sectional methods MSmetrix and SIENA(X) on test–retest scans from all MS patients in data set 1. Boxplots show absolute values of the whole brain (left) and gray matter (right) percentual volume change, computed either by the longitudinal approaches or based on two cross‐sectional measurements on the test–retest scans. Second row: Per‐scanner comparison of the measurement error of the longitudinal methods MSmetrix and SIENA on test–retest scans from all MS patients in data set 1. Boxplots show absolute values of the whole brain (left) and gray matter (right) percentual volume change
© Copyright Policy - creativeCommonsBy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5036437&req=5

brb3518-fig-0004: First row: Comparison of the measurement error of the longitudinal and cross‐sectional methods MSmetrix and SIENA(X) on test–retest scans from all MS patients in data set 1. Boxplots show absolute values of the whole brain (left) and gray matter (right) percentual volume change, computed either by the longitudinal approaches or based on two cross‐sectional measurements on the test–retest scans. Second row: Per‐scanner comparison of the measurement error of the longitudinal methods MSmetrix and SIENA on test–retest scans from all MS patients in data set 1. Boxplots show absolute values of the whole brain (left) and gray matter (right) percentual volume change

Mentions: In Fig. 4 the measurement error of the cross‐sectional methods is compared with those of the longitudinal methods on MS patients (data set 1), indicating the necessity of longitudinal measurements for reliable atrophy quantifications. Moreover, MSmetrix shows measurement errors comparable to or lower than SIENA(X). The test–retest percentage whole‐brain volume changes computed by MSmetrix‐long differ in absolute value from the expected 0% by 0.13% (median over all scan pairs, on all three scanners; first and third quartiles: 0.09–0.29%, maximum value: 0.7%), while those of SIENA differ from 0 in absolute value by 0.17% (first and third quartiles: 0.08–0.22%, maximum value: 1.2%). However, the difference between MSmetrix‐long and SIENA is not significant (p = .54 for the paired t‐test and p = .60 for the Wilcoxon signed‐rank test). For the cross‐sectional methods, the median percentage whole‐brain volume change error is 0.62% (first and third quartiles: 0.23–1.3%, maximum value: 3.8%) for MSmetrix‐cross and 0.82% (first and third quartiles: 0.34–2.04%, maximum value: 6.8%) for SIENAX. Also, the cross‐sectional methods MSmetrix‐cross and SIENAX are not significantly different (p = .10 for the paired t‐test and p = .16 for the Wilcoxon signed‐rank). However, MSmetrix‐long is significantly different from the cross‐sectional methods, that is MSmetrix‐cross and SIENAX (p < .01 for the paired t‐test and for the Wilcoxon signed‐rank test).


Reliable measurements of brain atrophy in individual patients with multiple sclerosis
First row: Comparison of the measurement error of the longitudinal and cross‐sectional methods MSmetrix and SIENA(X) on test–retest scans from all MS patients in data set 1. Boxplots show absolute values of the whole brain (left) and gray matter (right) percentual volume change, computed either by the longitudinal approaches or based on two cross‐sectional measurements on the test–retest scans. Second row: Per‐scanner comparison of the measurement error of the longitudinal methods MSmetrix and SIENA on test–retest scans from all MS patients in data set 1. Boxplots show absolute values of the whole brain (left) and gray matter (right) percentual volume change
© Copyright Policy - creativeCommonsBy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5036437&req=5

brb3518-fig-0004: First row: Comparison of the measurement error of the longitudinal and cross‐sectional methods MSmetrix and SIENA(X) on test–retest scans from all MS patients in data set 1. Boxplots show absolute values of the whole brain (left) and gray matter (right) percentual volume change, computed either by the longitudinal approaches or based on two cross‐sectional measurements on the test–retest scans. Second row: Per‐scanner comparison of the measurement error of the longitudinal methods MSmetrix and SIENA on test–retest scans from all MS patients in data set 1. Boxplots show absolute values of the whole brain (left) and gray matter (right) percentual volume change
Mentions: In Fig. 4 the measurement error of the cross‐sectional methods is compared with those of the longitudinal methods on MS patients (data set 1), indicating the necessity of longitudinal measurements for reliable atrophy quantifications. Moreover, MSmetrix shows measurement errors comparable to or lower than SIENA(X). The test–retest percentage whole‐brain volume changes computed by MSmetrix‐long differ in absolute value from the expected 0% by 0.13% (median over all scan pairs, on all three scanners; first and third quartiles: 0.09–0.29%, maximum value: 0.7%), while those of SIENA differ from 0 in absolute value by 0.17% (first and third quartiles: 0.08–0.22%, maximum value: 1.2%). However, the difference between MSmetrix‐long and SIENA is not significant (p = .54 for the paired t‐test and p = .60 for the Wilcoxon signed‐rank test). For the cross‐sectional methods, the median percentage whole‐brain volume change error is 0.62% (first and third quartiles: 0.23–1.3%, maximum value: 3.8%) for MSmetrix‐cross and 0.82% (first and third quartiles: 0.34–2.04%, maximum value: 6.8%) for SIENAX. Also, the cross‐sectional methods MSmetrix‐cross and SIENAX are not significantly different (p = .10 for the paired t‐test and p = .16 for the Wilcoxon signed‐rank). However, MSmetrix‐long is significantly different from the cross‐sectional methods, that is MSmetrix‐cross and SIENAX (p < .01 for the paired t‐test and for the Wilcoxon signed‐rank test).

View Article: PubMed Central - PubMed

ABSTRACT

Introduction: As neurodegeneration is recognized as a major contributor to disability in multiple sclerosis (MS), brain atrophy quantification could have a high added value in clinical practice to assess treatment efficacy and disease progression, provided that it has a sufficiently low measurement error to draw meaningful conclusions for an individual patient.

Metrixmetrix: In this paper, we present an automated longitudinal method based on Jacobian integration for measuring whole&#8208;brain and gray matter atrophy based on anatomical magnetic resonance images (MRI), named MS. MS is specifically designed to measure atrophy in patients with MS, by including iterative lesion segmentation and lesion filling based on FLAIR and T1&#8208;weighted MRI scans.

Metrixmetrixmetrixmetrix: MS is compared with SIENA with respect to test&ndash;retest error and consistency, resulting in an average test&ndash;retest error on an MS data set of 0.13% (MS) and 0.17% (SIENA) and a consistency error of 0.07% (MS) and 0.05% (SIENA). On a healthy subject data set including physiological variability the test&ndash;retest is 0.19% (MS) and 0.31% (SIENA).

Metrix: Therefore, we can conclude that MS could be of added value in clinical practice for the follow&#8208;up of treatment and disease progression in MS patients.

No MeSH data available.