National Institute of Neurological Disorders and Stroke
Translational Neuroradiology Section

Author Of 2 Presentations

Imaging Oral Presentation

FC03.02 - A step forward toward the fully automated assessment of the central vein sign

Speakers
Presentation Number
FC03.02
Presentation Topic
Imaging
Lecture Time
13:12 - 13:24

Abstract

Background

A deep-learning prototype method, called CVSNet, was recently introduced for the automated detection of the central vein sign (CVS) in brain lesions and demonstrated effective and accurate discrimination of multiple sclerosis (MS) from its mimics. However, this method solely considered focal lesions displaying the central vein sign (CVS+) or not (CVS), therefore requiring a manual pre-selection of the lesions to be evaluated by eliminating the so-called excluded lesions (CVSe) as defined by the NAIMS criteria. CVSe lesions may however play an important role in differential diagnosis. Moreover, extending the automated CVS classification to these lesions would facilitate the integration of CVSNet with existing MS lesion segmentation algorithms in a fully automated pipeline.

Objectives

To develop an improved version of the CVSNet prototype method able to classify all types of lesions (CVS+, CVS and CVSe).

Methods

Patients with an established MS or CIS diagnosis (RRMS 29; SPMS 10; PPMS 10; CIS 1; mean ± SD age: 50 ± 11 years; male/female: 23/27), and healthy controls (n=8; mean ± SD age: 41 ± 9 years; male/female: 5/3), underwent 3T brain MRI (MAGNETOM Skyra and MAGNETOM Prisma, Siemens Healthcare, Erlangen, Germany, or Achieva, Philips Healthcare, Best, Netherlands). Brain lesions were automatically segmented and manually corrected by a single rater. CVS assessment was conducted on FLAIR* images by two raters, according to the NAIMS guidelines, yielding 1542 CVS+, 1004 CVS−, and 1131 CVSe lesions. A convolutional neural network (CNN) based on the CVSnet architecture was trained with different configurations using 3021 samples (1261 CVS+, 847 CVS, and 913 CVSe) and evaluated in 656 unseen samples (281 CVS+, 157 CVS−, and 218 CVSe, from 13 patients) for final testing. The configurations relied on different combinations of the following channels as input: (i) FLAIR*, (ii) T2*, (iii) lesion mask, and (iv) CSF and brain tissue concentration maps obtained from a partial-volume estimation algorithm. Lesion-wise classification performance was evaluated for the different configurations by estimating the sensitivity, specificity, and accuracy for each lesion class.

Results

The results were similar across the different configurations. The best performance in the unseen testing set was obtained when all channels were used as input (sensitivity: 0.71, 0.73; specificity: 0.71, 0.81; and accuracy: 0.71, 0.79 for CVS+, CVS−, respectively). For CVSe, this approach achieved 0.52 sensitivity, 0.94 specificity, and 0.80 accuracy.

Conclusions

We introduced a modified CVSNet prototype method that can analyze the presence of the central vein for all types of brain lesions, enabling its integration with current MS lesion segmentation algorithms. This new feature will allow a fully automated assessment of the CVS in patients’ brains, speeding up the evaluation of CVS as a diagnostic biomarker for differentiating MS from mimicking diseases.

Collapse
Imaging Oral Presentation

PS11.05 - Inclusion of small ovoid lesions in central vein sign assessment improves sensitivity for multiple sclerosis

Speakers
Presentation Number
PS11.05
Presentation Topic
Imaging
Lecture Time
10:09 - 10:21

Abstract

Background

The ‘central vein sign’ (CVS) is increasingly recognized as a valuable biomarker with high specificity and sensitivity for multiple sclerosis (MS) MRI lesions. Current consensus North American Imaging in Multiple Sclerosis (NAIMS) guidelines recommend excluding lesions <3mm in diameter in any plane for CVS assessment. However, different lesion-size exclusion cut-offs for CVS have not been systematically evaluated.

Objectives

To evaluate the impact of different lesion size cut-offs and exclusion methodologies on CVS analysis and select3* criteria for MS diagnosis.

Methods

MS patients and non-MS controls were recruited as part of the National Institute of Neurological Disorders and Stroke MS natural history study and underwent 3T MRI on either Siemens Skyra or Philips Achieva scanners. MS lesions were segmented using a deep learning-based method and manually corrected by a single rater. Individual lesions were extracted as clusters of connected voxels, and their principal axes lengths (calculated as the lengths of the major axes of an ellipsoid with the same normalized second central moments) were used to measure lesion size in 3 dimensions. Ground truth CVS assessment was conducted by two raters on all lesions regardless of size. Two paradigms of lesion exclusion were compared: (1) excluding lesions if any dimension was less than threshold (ExcAny), or (2) if all dimensions were less than threshold (ExcAll).

Results

A total of 3920 lesions from 71 subjects (8 healthy controls, 36 RRMS, 12 SPMS, 14 PPMS, 1 CIS) were included in the analysis. CVS+ lesions were more likely to be ovoid and less spherical compared to their CVS- counterparts, as measured by the fractional anisotropy of lesion dimensions (mean difference 0.02, p=0.001). Of the 1679 CVS+ lesions in the cohort, 82% met the ExcAny criteria to be excluded at a 3mm cut-off, which was reduced to 29% when ExcAll criteria were used (McNemar test, p < 0.001). At the subject-level, an increase in the sensitivity of select3* CVS criteria for MS diagnosis was noted at 3mm using the less strict ExcAll (95%) compared to the more conservative ExcAny criteria (61%), without impacting specificity (100% for both methods). There was a reduction in specificity for both ExcAny and ExcAll criteria when size cut-offs less than or equal to 2mm were used (88% for both).

Conclusions

Compared to the current NAIMS guidelines, ExcAll criteria for CVS lesion analysis allow the inclusion of a larger proportion of CVS+ lesions and improve the sensitivity of select3* criteria for MS diagnosis. These findings improve the applicability of the CVS as a diagnostic marker for MS in clinical practice and provide evidence for future modifications of CVS lesion exclusion guidelines.

Collapse

Author Of 2 Presentations

Biomarkers and Bioinformatics Poster Presentation

P0083 - Gadolinium improves detection of central vein lesions in MS using 3T FLAIR*. (ID 1404)

Abstract

Background

The central vein sign (CVS) is a proposed MRI diagnostic biomarker for multiple sclerosis (MS). Use of gadolinium (Gd) in the CVS literature has been inconsistent, and it is unknown whether Gd improves detection of CVS when using FLAIR*.

Objectives

To determine if, and to what extent, gadolinium injection improves detection of CVS lesions when using FLAIR* imaging.

Methods

A cross-sectional multicenter study recruited adults clinically and/or radiologically suspected of having MS. High-isotropic-resolution, T2*-weighted segmented echo-planar imaging (T2*-EPI) was acquired pre- and post-injection of Gd-based contrast agent at 3T; pre-Gd 3D FLAIR images were also acquired. T2*-EPI and FLAIR images were processed on the QMENTA platform to generate FLAIR* images. FLAIR* pre-Gd and post-Gd scans from this substudy of 30 patients at 5 sites were analyzed. FLAIR images were used to create T2 lesion masks. Subsequently, FLAIR* images were evaluated in a random order. Lesions were categorized as CVS+, CVS-, or excluded based on the North American Imaging in MS (NAIMS) Criteria by two trained raters blinded to clinical data and Gd use. The proportion of CVS+ lesions was calculated for each scan, and differences in CVS detection based on Gd use were assessed by a Wilcoxon rank-sum test. Diagnostic performance was compared against McDonald 2017 Criteria.

Results

The mean participant age was 45 years (SD: 12); 23 (77%) were women. 14 (47%) met McDonald 2017 Criteria for MS, while 16 (53%) did not (“non-MS”). A total of 487 CVS+ lesions and 976 CVS- lesions were evaluated. The percentage of CVS+ lesions post-Gd in the MS group (median 67% [IQR 30%]) was higher than pre-Gd (41% [47%], p<0.001). There was no apparent difference in percentage of CVS+ lesion in the non-MS group (post-Gd: 10% [23%]; pre-Gd: 5% [29%]; p=0.1). In the MS group, 12/14 (86%) had ≥40% CVS+ lesions on post-Gd imaging, whereas only 8/14 (57%) exceeded that threshold on pre-Gd imaging. When evaluating CVS performance using the 40% CVS+ threshold, the sensitivity and specificity of the CVS post-Gd for MS were 86% and 81%, respectively, compared to 54% and 86% pre-Gd.

Conclusions

The detection of the CVS using FLAIR* at 3T is improved when Gd is used. Based on these results, a multicenter prospective CVS diagnostic study, sponsored by NINDS and NAIMS, will use Gd in the study protocol. Future clinical use of the CVS should balance the increased costs and potential risks of Gd use with the risks of misdiagnosis due to missing CVS on non-contrast imaging.

Collapse
Imaging Poster Presentation

P0542 - Assessment of central vein sign conspicuity in multicenter 3T FLAIR* imaging (ID 985)

Abstract

Background

The central vein sign (CVS) is a proposed diagnostic biomarker for MS that can be identified using FLAIR*. The robustness of 3T FLAIR*, with and without the injection of gadolinium contrast agent (Gd), for imaging the CVS in a multicenter setting has not yet been demonstrated.

Objectives

To assess the conspicuity of the CVS on 3T FLAIR* imaging acquired across different sites with and without the injection of Gd.

Methods

A cross-sectional multicenter study recruited adults with a clinical and/or radiological suspicion of having MS from 10 sites within the North American Imaging in MS (NAIMS) Cooperative. High-isotropic-resolution T2*-weighted segmented echo-planar imaging (T2*-EPI) was acquired at 3T, pre- and post-injection of Gd, along with 3D FLAIR on different scanner brands and models. T2*-EPI and FLAIR images were processed on an online imaging platform (QMENTA) to generate FLAIR* images. To objectively assess the conspicuity of the CVS inside MS lesions, lesions and veins were segmented automatically and used to compute lesion-to-vein contrast-to-noise ratio (CNR) measures. ANOVA was used to compare CNR values across sites with post-hoc Tukey Honest Significant Difference testing. Multiple testing between sites was considered by controlling the false discovery rate. One-sided paired t-testing was used to compare the overall lesion-to-vein CNR values between pre- and post-Gd FLAIR*.

Results

Seventy-eight patients from nine sites were included in the analysis; one site was excluded due to low enrollment. The overall mean(coefficient of variation, CV) lesion-to-vein CNR values across the nine sites were 0.35(14%) and 0.37(12%) for pre- and post-Gd FLAIR*, respectively. Excluding an additional site that used an unharmonized FLAIR acquisition, the resulting mean(CV) CNR values were 0.36(12%) for pre-Gd and 0.37(11%) for post-Gd FLAIR*. Across most sites, there was a significant improvement in lesion-to-vein CNR measures for post-Gd compared to pre-Gd FLAIR* [mean difference = 0.011, p < 0.001, 95% CI: (0.008,0.015)].

Conclusions

Lesion-to-vein CNR measures across sites are in line with values first published for 3T FLAIR* and demonstrate the robustness of 3T FLAIR* for imaging the CVS in a multicenter setting. Moreover, there was an increase in vein conspicuity with improvement in CNR on post-Gd FLAIR*. Based on these results, a prospective multicenter NAIMS CVS diagnostic study, sponsored by NINDS, will use 3T FLAIR* imaging with Gd in the study protocol.

Collapse

Presenter Of 1 Presentation

Biomarkers and Bioinformatics Poster Presentation

P0083 - Gadolinium improves detection of central vein lesions in MS using 3T FLAIR*. (ID 1404)

Abstract

Background

The central vein sign (CVS) is a proposed MRI diagnostic biomarker for multiple sclerosis (MS). Use of gadolinium (Gd) in the CVS literature has been inconsistent, and it is unknown whether Gd improves detection of CVS when using FLAIR*.

Objectives

To determine if, and to what extent, gadolinium injection improves detection of CVS lesions when using FLAIR* imaging.

Methods

A cross-sectional multicenter study recruited adults clinically and/or radiologically suspected of having MS. High-isotropic-resolution, T2*-weighted segmented echo-planar imaging (T2*-EPI) was acquired pre- and post-injection of Gd-based contrast agent at 3T; pre-Gd 3D FLAIR images were also acquired. T2*-EPI and FLAIR images were processed on the QMENTA platform to generate FLAIR* images. FLAIR* pre-Gd and post-Gd scans from this substudy of 30 patients at 5 sites were analyzed. FLAIR images were used to create T2 lesion masks. Subsequently, FLAIR* images were evaluated in a random order. Lesions were categorized as CVS+, CVS-, or excluded based on the North American Imaging in MS (NAIMS) Criteria by two trained raters blinded to clinical data and Gd use. The proportion of CVS+ lesions was calculated for each scan, and differences in CVS detection based on Gd use were assessed by a Wilcoxon rank-sum test. Diagnostic performance was compared against McDonald 2017 Criteria.

Results

The mean participant age was 45 years (SD: 12); 23 (77%) were women. 14 (47%) met McDonald 2017 Criteria for MS, while 16 (53%) did not (“non-MS”). A total of 487 CVS+ lesions and 976 CVS- lesions were evaluated. The percentage of CVS+ lesions post-Gd in the MS group (median 67% [IQR 30%]) was higher than pre-Gd (41% [47%], p<0.001). There was no apparent difference in percentage of CVS+ lesion in the non-MS group (post-Gd: 10% [23%]; pre-Gd: 5% [29%]; p=0.1). In the MS group, 12/14 (86%) had ≥40% CVS+ lesions on post-Gd imaging, whereas only 8/14 (57%) exceeded that threshold on pre-Gd imaging. When evaluating CVS performance using the 40% CVS+ threshold, the sensitivity and specificity of the CVS post-Gd for MS were 86% and 81%, respectively, compared to 54% and 86% pre-Gd.

Conclusions

The detection of the CVS using FLAIR* at 3T is improved when Gd is used. Based on these results, a multicenter prospective CVS diagnostic study, sponsored by NINDS and NAIMS, will use Gd in the study protocol. Future clinical use of the CVS should balance the increased costs and potential risks of Gd use with the risks of misdiagnosis due to missing CVS on non-contrast imaging.

Collapse