Vadim Malis¹, Jirach Kungsamutr², Won Bae¹, Xiaowei Zhang¹, Yoshimori Kassai³, Albert Hsiao¹, John S Lane⁴, and Mitsue Miyazaki^1
1Radiology, UC San Diego, San Diego, CA, United States, ²Bioengineering, UC San Diego, San Diego, CA, United States, ³Canon Medical Systems Corp., Tochigi, Japan, ⁴Surgery, UC San Diego, San Diego, CA, United States

Iliac vein compression syndrome, known as May-Thurner syndrome (MTS), is a condition when the right common iliac artery constrains the flow in the left common iliac vein. Currently, ultrasound and CT venography (CTV) are the imaging modalities of choice for the diagnosis. We propose a framework to visualize iliac vein and artery utilizing contrast free MR venography and MR angiography in less than 10 minutes.

Huiming Dong¹, Henrik Haraldsson¹, Joseph Leach¹, Ang Zhou¹, Megan Ballweber¹, Chengcheng Zhu², Yue Xuan³, Zhongjie Wang³, Michael Hope¹, Liang Ge³, Frederick H. Epstein⁴, David Saloner¹, Elaine Tseng³, and Dimitrios Mitsouras^1
1Department of Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, CA, United States, ²Department of Radiology, University of Washington, Seattle, WA, United States, ³Department of Surgery, University of California, San Francisco, San Francisco, CA, United States, ⁴Department of Biomedical Engineering, University of Virginia, Charlottesville, VA, United States

Ascending thoracic aortic aneurysm (aTAA) can result in life-threatening rupture or dissection. Displacement encoding with stimulated echoes (DENSE) is a non-invasive phase-contrast MRI technique that can measure aTAA wall deformation during the cardiac cycle. This study investigated DENSE-derived aTAA wall stretch in patients and found that the ratio between aTAA stretch and descending aorta stretch was different in patients who met surgical repair criteria from those who did not. Moreover, mechanical properties of aTAA specimens from patients who underwent surgery correlated significantly with in vivo DENSE measurements. Our findings suggest DENSE as a potential imaging marker for understanding aTAA progression.

Mohammed Salman Shazeeb¹, Robert King¹, Zeynep Vardar¹, Josephine Kolstad¹, Vania Anagnostakou¹, Nils Henninger¹, and Matthew Gounis^1
1University of Massachusetts Medical School, Worcester, MA, United States

In acute ischemic stroke due to large vessel occlusion (LVO), information about the penumbral tissue can be vital in making decisions on how to treat ischemic stroke patients in the clinic. This study investigated the use of intravoxel incoherent motion (IVIM) MRI in a canine LVO model to quantify the perfusion information in penumbral tissue prior to infarction. The IVIM parameters were assessed to predict the onset of penumbral tissue death. Of the 3 IVIM parameters, f showed the best utility in predicting penumbral tissue death.

Jiayu Xiao¹, Jae W Song², Steven Y Cen^1,3, Fang Wu⁴, Xiao Liu⁵, Tao Jiang⁵, Konrad H Schlick⁶, Debiao Li⁷, Shlee S Song⁶, Qi Yang⁵, and Zhaoyang Fan^1,8,9
1Radiology, Keck School of Medicine, Los Angeles, CA, United States, ²Radiology, Hospital of the University of Pennsylvania, Philadelphia, PA, United States, ³Neurology, Keck School of Medicine, Los Angeles, CA, United States, ⁴Radiology, Xuanwu Hospital, Beijing, China, ⁵Radiology, Beijing Chaoyang Hospital, Beijing, China, ⁶Neurology, Cedars-Sinai Medical Center, Los Angeles, CA, United States, ⁷Biomedical Imaging Research Institute, Cedars-Sinai Medical Center, Los Angeles, CA, United States, ⁸Radiation Oncology, Keck School of Medicine, Los Angeles, CA, United States, ⁹Biomedical Engineering, University of Southern California, Los Angeles, CA, United States

 Hypertension is one of the most important modifiable risk factor for atherosclerosis. However, the association between blood pressure and intracranial plaque remains unclear. In this study, we investigated the impact of hypertension on intracranial plaque burden and distribution in 174 patients with recent ischemic stroke. Systolic blood pressure and diastolic blood pressure were independently associated with a higher total plaue burden as well as higher plaque counts at the proximal and junctional locations. These findings suggest hypertension may influence not only the count burden but also the distribution of intracranial plaque development.

Samantha E Seymour¹, Ryan A Rava¹, Mitchell T Chudzik¹, Kenneth V Snyder^1,2,3, Muhammad E Waqas^1,3, Jason M Davies^1,2,3,4, Elad E Levy^1,2,3, Adnan E Siddiqui^1,2,3, Xiaoliang E Zhang⁵, and Ciprian E Ionita^1,2,3,6
1Canon Stroke and Vascular Center, Buffalo, NY, United States, ²Jacobs School of Medicine and Biomedical Sciences, Buffalo, NY, United States, ³Department of Neurosurgery, University at Buffalo, Buffalo, NY, United States, ⁴Department of Bioinformatics, University at Buffalo, Buffalo, NY, United States, ⁵Department of Biomedical Engineering, University at Buffalo, Buffalo, NY, United States, ⁶Department of Biomedical Engineerinng, University at Buffalo, Buffalo, NY, United States

Intracranial hemorrhage (ICH) is bleeding within the cranium and occurs within the brain tissue, ventricles, and intracranial space. Hematoma expansion following an ICH has been related to increased mortality and morbidity inpatients. To detect ICH patients at risk, machine learning models can be used to predict whether or not hematoma expansion will occur. This study aims to assess the feasibility of machine learning prediction models using a radiomics approach. The highest sensitivity results indicated as 95% confidence intervals are 0.68 ± 0.004 and 0.72 ± 0.004, were achieved by support vector machine and logistic regression classifier models, respectively.

Li Jiang¹, Yongsheng Chen², Marco Muccio¹, Chenyang Li¹, Sagar Buch², E Mark Haacke², and Yulin Ge^1
1NYU Grossman School of Medicine, New York, NY, United States, ²Wayne State University, Detroit, MI, United States

Ultra-small superparamagnetic iron oxides (USPIO) enhanced MRI has shown its great potential in detecting vascular abnormality in the brain on T2*-weighted sequences for better showing both macro- and micro-vasculature. In this study, we investigated the quantitative phase changes after Ferumoxytol infusion and examined its residual contrast in brain vasculature at 3-, 7- and 14-day follow-up SWI scans. To quantitatively measure the Ferumoxytol-induced susceptibility, the veins parallel to the main field were chosen for measurement. After measuring the intensity on high-pass filter phase images, the paired t-test showed a significant difference between pre-contrast and post-3d values from the follow-up visits.

Hanyue Zhou^1,2, Jiayu Xiao², Dan Ruan^1,3, and Zhaoyang Fan^2,4,5
1Bioengineering, University of California, Los Angeles, Los Angeles, CA, United States, ²Radiology, University of Southern California, Los Angeles, CA, United States, ³Radiation Oncology, University of California, Los Angeles, Los Angeles, CA, United States, ⁴Radiation Oncology, University of Southern California, Los Angeles, CA, United States, ⁵Biomedical Engineering, University of Southern California, Los Angeles, CA, United States

We propose an automatic pipeline for atherosclerosis plaque analysis based on magnetic resonance (MR) vessel wall imaging (VWI). The pipeline consists of five modules, i.e., 1) MR angiography (MRA) to VWI registration, 2) vessel centerline tracking, 3) vessel straightening and cross-sectional slices generating, 4) deep learning -based vessel wall and lumen segmentation, and 5) plaque quantification. The pipeline achieved good to excellent reliability for the clinically relevant plaque measures and largely reduced the analysis time required for physicians.

Hanyue Zhou^1,2, Jiayu Xiao², Dan Ruan^1,3, and Zhaoyang Fan^2,4,5
1Bioengineering, University of California, Los Angeles, Los Angeles, CA, United States, ²Radiology, University of Southern California, Los Angeles, CA, United States, ³Radiation Oncology, University of California, Los Angeles, Los Angeles, CA, United States, ⁴Radiation Oncology, University of Southern California, Los Angeles, CA, United States, ⁵Biomedical Engineering, University of Southern California, Los Angeles, CA, United States

With a typical slice-by-slice labeling fashion, the manual contouring process is subject to large intra-observer variation, especially for small-sized intracranial arteries. We propose an iterative refinement approach for ground truth contours with the help of deep neural networks for intracranial lumen and vessel wall segmentations. We demonstrate that the approach improved the smoothness of the predicted contours and feature quantification, which can potentially boost the robustness of a neural network as a consequence of the  reduced uncertainty in expert labels.

Mona Sharifi Sarabi¹, Samantha J Ma^1,2, Danny JJ Wang¹, and Yonggang Shi^1
1University of Southern California, Los Angeles, CA, United States, ²Siemens Medical Solutions USA, Inc., Los Angeles, CA, United States

High resolution 3D black blood MRI with sub-millimeter spatial resolution has been proposed for visualizing small vessels of the brain. Here we present and evaluate a novel approach for mapping brain vessel density from 3D black blood images acquired at 3T. Using automated vessel segmentation and non-linear registration, localized detection and quantification of small vessel density is demonstrated to be feasible.  This framework can potentially serve as a useful tool for detection and monitoring of localized vascular changes in aging and neurovascular disorders.

Ran Li^1,2, Jie Zheng^1,2, Pamela Woodard^1,2, and Jha Abhinav^1,2
1Mallinckrodt Institute of Radiology, Washington University in St. Louis, St. Louis, MO, United States, ²Department of Biomedical Engineering, Washington University in St. Louis, St. Louis, MO, United States

Our objective of this study is to accurately segment and classify carotid atherosclerotic plaque components in a completely automated manner, based on multi-weighted MRI. Specifically, we segmented each pixel using a two-module neural network model. Furthermore, we generated segmentation uncertainty maps with a Bayesian method to evaluate the inherent uncertainty of this segmentation task.

John Cerne¹, Sophia Liu¹, Muhammad Umair¹, Ashitha Pathrose¹, Hatice Savas¹, Lisa Wilsbacher¹, Jackson E Moore¹, Michael Markl¹, James Carr¹, and Ryan Avery^1
1Northwestern University, Chicago, IL, United States

Combined modality PET/MR may be a sensitive modality to detect severe large vessel vasculitis. To distinguish severe large vessel vasculitis, a qualitative review of PET imaging findings appears more useful than standardized uptake values of a patient’s highest-uptake vessel. Inflammatory marker levels and trends, as well as single-modality scan findings, were not significantly different between severe and non-severe large vessel vasculitis patients.

Seong-Eun Kim¹, John A Roberts¹, J. Scott MacNally¹, Mathew Alexander ¹, Gerald S Treiman^2,3, Hediyeh Baradaran¹, and Dennis L Parker^1
1UCAIR, Department of Radiology and Imaging Sciences, University of Utah, Salt Lake City, UT, United States, ²Department of Surgery, University of Utah, Salt Lake City, UT, United States, ³VASLCHCS, Salt Lake City, UT, United States

This work introduced a reliable, comprehensive carotid DCE technique, including a high spatial and temporal resolution 3D dynamic T1 sequence and retrospective reconstruction methods to measure kinetic parameters on carotid atherosclerotic plaque. This technique can be used to evaluate the correlation between kinetic parameters and ischemic stroke. We believe that plaque kinetic parameters can be used in subsequent clinical trials as a changeable predictor of ischemic stroke risk. Clinicians could then use carotid DCE to monitor early response to medical therapy, and thereby improve detection and treatment of unstable carotid plaque and prevention of stroke.