Multi-Planar, Multi-Contrast and Multi-Time Point Analysis Tool (MOCHA) for Intracranial Vessel Wall Characterization

Automated MRI-based segmentation of intracranial arterial calcification by restricting feature complexity

PURPOSE

To develop an automated deep learning model for MRI-based segmentation and detection of intracranial arterial calcification.

METHODS

A novel deep learning model under the variational autoencoder framework was developed. A theoretically grounded dissimilarity loss was proposed to refine network features extracted from MRI and restrict their complexity, enabling the model to learn more generalizable MR features that enhance segmentation accuracy and robustness for detecting calcification on MRI.

RESULTS

The proposed method was compared with nine baseline methods on a dataset of 113 subjects and showed superior performance (for segmentation, Dice similarity coefficient: 0.620, area under precision-recall curve [PR-AUC]: 0.660, 95% Hausdorff Distance: 0.848 mm, Average Symmetric Surface Distance: 0.692 mm; for slice-wise detection, F1 score: 0.823, recall: 0.764, precision: 0.892, PR-AUC: 0.853). For clinical needs, statistical tests confirmed agreement between the true calcification volumes and predicted values using the proposed approach. Various MR sequences, namely T1, time-of-flight, and SNAP, were assessed as inputs to the model, and SNAP provided unique and essential information pertaining to calcification structures.

CONCLUSION

The proposed deep learning model with a dissimilarity loss to reduce feature complexity effectively improves MRI-based identification of intracranial arterial calcification. It could help establish a more comprehensive and powerful pipeline for vascular image analysis on MRI.

Simulation-based optimization and experimental comparison of intracranial T2-weighted DANTE-SPACE vessel wall imaging at 3T and 7T

PURPOSE

T2-weighted DANTE-SPACE (Delay Alternating with Nutation for Tailored Excitation - Sampling Perfection with Application optimized Contrasts using different flip angle Evolution) sequences facilitate non-invasive intracranial vessel wall imaging at 7T through simultaneous suppression of blood and CSF. However, the achieved vessel wall delineation depends closely on the selected sequence parameters, and little information is available about the performance of the sequence using more widely available 3T MRI. Therefore, in this paper a comprehensive DANTE-SPACE simulation framework is used for the optimization and quantitative comparison of T2-weighted DANTE-SPACE at both 7T and 3T.

METHODS

Simulations are used to propose optimized sequence parameters at both 3T and 7T. At 7T, an additional protocol which uses a parallel transmission (pTx) shim during the DANTE preparation for improved suppression of inflowing blood is also proposed. Data at both field strengths using optimized and literature protocols are acquired and quantitatively compared in six healthy volunteers.

RESULTS

At 7T, more vessel wall signal can be retained while still achieving sufficient CSF suppression by using fewer DANTE pulses than described in previous implementations. The use of a pTx shim during DANTE at 7T provides a modest further improvement to the inner vessel wall delineation. At 3T, aggressive DANTE preparation is required to achieve CSF suppression, resulting in reduced vessel wall signal. As a result, the achievable vessel wall definition at 3T is around half that of 7T.

CONCLUSION

Simulation-based optimization of DANTE parameters facilitates improved T2-weighted DANTE-SPACE contrasts at 7T. The improved vessel definition of T2-weighted DANTE-SPACE at 7T makes DANTE preparation more suitable for T2-weighted VWI at 7T than at 3T.

In Vivo Deformation of the Human Basilar Artery

An estimated 6.8 million people in the United States have an unruptured intracranial aneurysms, with approximately 30,000 people suffering from intracranial aneurysms rupture each year. Despite the development of population-based scores to evaluate the risk of rupture, retrospective analyses have suggested the limited usage of these scores in guiding clinical decision-making. With recent advancements in imaging technologies, artery wall motion has emerged as a promising biomarker for the general study of neurovascular mechanics and in assessing the risk of intracranial aneurysms. However, measuring arterial wall deformations in vivo itself poses several challenges, including how to image local wall motion and deriving the anisotropic wall strains over the cardiac cycle. To overcome these difficulties, we first developed a novel in vivo MRI-based imaging method to acquire cardiac gated images of the human basilar artery (BA) over the cardiac cycle. Next, complete BA endoluminal surfaces from each frame were segmented, producing high-resolution point clouds of the endoluminal surfaces. From these point clouds we developed a novel B-spline-based surface representation, then exploited the local support nature of B-splines to determine the local endoluminal surface strains. Results indicated distinct regional and temporal variations in BA wall deformation, highlighting the heterogeneous nature BA function. These included large circumferential strains (up to ∼ 20 % ), and small longitudinal strains, which were often contractile and out of phase with the circumferential strains patterns. Of particular interest was the temporal phase lag in the maximum circumferential perimeter length, which indicated that the BA deforms asynchronously over the cardiac cycle. In summary, the proposed method enabled local deformation analysis, allowing for the successful reproduction of local features of the BA, such as regional principal stretches, areal changes, and pulsatile motion. Integrating the proposed method into existing population-based scores has the potential to improve our understanding of mechanical properties of human BA and enhance clinical decision-making.

Plaque Evolution and Vessel Wall Remodeling of Intracranial Arteries: A Prospective, Longitudinal Vessel Wall MRI Study

BACKGROUND

Progression of intracranial atherosclerotic disease (ICAD) is associated with ischemic stroke events and can be quantified with three-dimensional (3D) intracranial vessel wall (IVW) MRI. However, longitudinal 3D IVW studies are limited and ICAD evolution remains relatively unknown.

PURPOSE

To evaluate ICAD changes longitudinally and to characterize the imaging patterns of atherosclerotic plaque evolution.

STUDY TYPE

Prospective.

POPULATION

37 patients (69 ± 12 years old, 12 females) with angiography confirmed ICAD.

FIELD STRENGTH/SEQUENCE

3.0T/3D time-of-flight gradient echo sequence and T1- and proton density-weighted fast spin echo sequences.

ASSESSMENT

Each patient underwent baseline and 1-year follow-up IVW. Then, IVW data from both time points were jointly preprocessed using a multitime point, multicontrast, and multiplanar viewing workflow (known as MOCHA). Lumen and outer wall of plaques were traced and measured, and plaques were then categorized into progression, stable, and regression groups based on changes in plaque wall thickness. Patient demographic and clinical data were collected. Culprit plaques were identified based on cerebral ischemic infarcts.

STATISTICAL TESTS

Generalized estimating equations-based linear and logistic regressions were used to assess associations between vascular risk factors, medications, luminal stenosis, IVW plaque imaging features, and longitudinal changes. A two-sided P-value<0.05 was considered statistically significant.

RESULTS

Diabetes was significantly associated with ICAD progression, resulting in 6.6% decrease in lumen area and 6.7% increase in wall thickness at 1-year follow-up. After accounting for arterial segments, baseline contrast enhancement predicted plaque progression (odds ratio = 3.61). Culprit plaques experienced an average luminal expansion of 10.9% after 1 year. 74% of the plaques remained stable during follow-up. The regression group (18 plaques) showed significant increase in minimum lumen area (from 7.4 to 8.3 mm2), while the progression group (13 plaques) showed significant decrease in minimum lumen area (from 5.4 to 4.3 mm2).

DATA CONCLUSION

Longitudinal 3D IVW showed ICAD remodeling on the lumen side. Culprit plaques demonstrated longitudinal luminal expansion compared with their non-culprit counterparts. Baseline plaque contrast enhancement and diabetes mellitus were found to be significantly associated with ICAD changes.

EVIDENCE LEVEL

2 TECHNICAL EFFICACY: Stage 3.

An extended phase graph-based framework for DANTE-SPACE simulations including physiological, temporal, and spatial variations

PURPOSE

The delay alternating with nutation for tailored excitation (DANTE)-sampling perfection with application-optimized contrasts (SPACE) sequence facilitates 3D intracranial vessel wall imaging with simultaneous suppression of blood and CSF. However, the achieved image contrast depends closely on the selected sequence parameters, and the clinical use of the sequence is limited in vivo by observed signal variations in the vessel wall, CSF, and blood. This paper introduces a comprehensive DANTE-SPACE simulation framework, with the aim of providing a better understanding of the underlying contrast mechanisms and facilitating improved parameter selection and contrast optimization.

METHODS

An extended phase graph formalism was developed for efficient spin ensemble simulation of the DANTE-SPACE sequence. Physiological processes such as pulsatile flow velocity variation, varying flow directions, intravoxel velocity variation, diffusion, andB 1 + $$ {\mathrm{B}}_1^{+} $$ effects were included in the framework to represent the mechanisms behind the achieved signal levels accurately.

RESULTS

Intravoxel velocity variation improved temporal stability and robustness against small velocity changes. Time-varying pulsatile velocity variation affected CSF simulations, introducing periods of near-zero velocity and partial rephasing. Inclusion of diffusion effects was found to substantially reduce the CSF signal. Blood flow trajectory variations had minor effects, butB 1 + $$ {\mathrm{B}}_1^{+} $$ differences along the trajectory reduced DANTE efficiency in low-B 1 + $$ {\mathrm{B}}_1^{+} $$ areas. Introducing low-velocity pulsatility of both CSF and vessel wall helped explain the in vivo observed signal heterogeneity in both tissue types.

CONCLUSION

The presented simulation framework facilitates a more comprehensive optimization of DANTE-SPACE sequence parameters. Furthermore, the simulation framework helps to explain observed contrasts in acquired data.

Vessel Wall Imaging in Cryptogenic Stroke

Cryptogenic strokes are symptomatic cerebral ischemic infarcts without a clear etiology identified following standard diagnostic evaluation and currently account for 10% to 40% of stroke cases. Continued research is needed to identify and bridge gaps in knowledge of this stroke grouping. Vessel wall imaging has increasingly shown its utility in the diagnosis and characterization of various vasculopathies. Initial promising evidence suggests rational use of vessel wall imaging in stroke workup may unravel pathologies that otherwise would have been occult and further improve our understanding of underlying disease processes that can translate into improved patient outcomes and secondary stroke prevention.

VWI-APP: Vessel wall imaging-dedicated automated processing pipeline for intracranial atherosclerotic plaque quantification

BACKGROUND

Quantitative plaque assessment based on 3D magnetic resonance (MR) vessel wall imaging (VWI) has been shown to provide valuable numerical markers of the burden and risk of intracranial atherosclerotic disease (ICAD). However, plaque quantification is currently time-consuming and observer-dependent due to the demand for heavy manual effort. A VWI-dedicated automated processing pipeline (VWI-APP) is desirable.

PURPOSE

To develop and evaluate a VWI-APP for end-to-end quantitative analysis of intracranial atherosclerotic plaque.

METHODS

We retrospectively enrolled 91 subjects with ICAD (80 for pipeline development, 10 for an end-to-end pipeline evaluation, and 1 for demonstrating longitudinal plaque assessment) who had undergone VWI and MR angiography. In an end-to-end evaluation, diameter stenosis (DS), normalized wall index (NWI), remodeling ratio (RR), plaque wall contrast ratio (CR), and total plaque volume (TPV) were quantified at each culprit lesion using the developed VWI-APP and a computer-aided manual approach by a neuroradiologist, respectively. The time consumed in each quantification approach was recorded. Two-sided paired t-tests and intraclass correlation coefficient (ICC) were used to determine the difference and agreement in each plaque metric between VWI-APP and manual quantification approaches.

RESULTS

There was no significant difference between VWI-APP and manual quantification in each plaque metric. The ICC was 0.890, 0.813, 0.827, 0.891, and 0.991 for DS, NWI, RR, CR, and TPV, respectively, suggesting good to excellent accuracy of the pipeline method in plaque quantification. Quantitative analysis of each culprit lesion on average took 675.7 s using the manual approach but shortened to 238.3 s with the aid of VWI-APP.

CONCLUSIONS

VWI-APP is an accurate and efficient approach to intracranial atherosclerotic plaque quantification. Further clinical assessment of this automated tool is warranted to establish its utility in the risk assessment of ICAD lesions.

Survey of the American Society of Neuroradiology Membership on the Use and Value of Extracranial Carotid Vessel Wall MRI

BACKGROUND AND PURPOSE

Extracranial vessel wall MRI (EC-VWI) contributes to vasculopathy characterization. This survey study investigated EC-VWI adoption by American Society of Neuroradiology (ASNR) members and indications and barriers to implementation.

MATERIALS AND METHODS

The ASNR Vessel Wall Imaging Study Group survey on EC-VWI use, frequency, applications, MR imaging systems and field strength used, protocol development approaches, vendor engagement, reasons for not using EC-VWI, ordering provider interest, and impact on clinical care was distributed to the ASNR membership between April 2, 2019, to August 30, 2019.

RESULTS

There were 532 responses; 79 were excluded due to minimal, incomplete response and 42 due to redundant institutional responses, leaving 411 responses. Twenty-six percent indicated that their institution performed EC-VWI, with 66.3% performing it ≤1-2 times per month, most frequently on 3T MR imaging, with most using combined 3D and 2D protocols. Protocols most commonly included pre- and postcontrast T1-weighted imaging, TOF-MRA, and contrast-enhanced MRA. Inflammatory vasculopathy (63.3%), plaque vulnerability assessments (61.1%), intraplaque hemorrhage (61.1%), and dissection-detection/characterization (51.1%) were the most frequent applications. For those not performing EC-VWI, the reasons were a lack of ordering provider interest (63.9%), lack of radiologist time/interest (47.5%) or technical support (41.4%) for protocol development, and limited interpretation experience (44.9%) and knowledge of clinical applications (43.7%). Reasons given by 46.9% were that no providers approached radiology with interest in EC-VWI. If barriers were overcome, 51.1% of those not performing EC-VWI indicated they would perform it, and 40.6% were unsure; 48.6% did not think that EC-VWI had impacted patient management at their institution.

CONCLUSIONS

Only 26% of neuroradiology groups performed EC-VWI, most commonly due to limited clinician interest. Improved provider and radiologist education, protocols, processing techniques, technical support, and validation trials could increase adoption.

Characterization of non-stenotic plaques in intracranial arteries with multi-contrast, multi-planar vessel wall image analysis

OBJECTIVES

Non-stenotic plaques have been observed in intracranial arteries but are less understood compared to those in coronary and carotid arteries. We sought to compare plaque distribution and morphology between stenotic and non-stenotic intracranial plaques with MR vessel wall imaging (VWI) and quantitative image analysis.

MATERIALS AND METHODS

Twenty-four patients with intracranial arterial stenosis or luminal irregularity on clinical imaging were scanned with a multi-contrast VWI protocol. Plaques were detected as focal wall thickening on co-registered multiplanar reformats of multi-contrast VWI, with assessment of the location and morphology. TOF-MRA was independently reviewed for any appreciable stenosis using the WAISD criteria.

RESULTS

Across 504 arterial segments, a total of 80 plaques were detected, including 23 (29%) with stenosis on TOF-MRA, 56 (70%) without, and 1 (1%) not covered by TOF-MRA. Plaques involving the ICA were more likely to be non-stenotic than those involving other segments (80% versus 55%, p = 0.030) whereas the basilar artery (40%) and PCA (33%) had the lowest proportions of non-stenotic plaques. Maximum wall thickness, indicative of plaque burden, correlated poorly with degree of stenosis (p = 0.10) and overlapped substantially between stenotic and non-stenotic plaques (1.9 [1.5, 2.4] versus 2.0 [1.5, 2.2] mm, p = 0.074).

CONCLUSIONS

Intracranial plaques without appreciable stenosis on TOF-MRA represent a large proportion of lesions throughout arterial segments but disproportionately affect the ICA. Morphological characterization of plaques with and without stenosis shows that luminal stenosis is a poor indicator of the underlying burden of intracranial atherosclerosis.