Je Yeong Sone¹, Yan Li^1,2, Nicholas Hobson¹, Sharbel G. Romanos¹, Abhinav Srinath¹, Seán B. Lyne¹, Abdallah Shkoukani¹, Julián Carrión-Penagos¹, Agnieszka Stadnik¹, Kristina Piedad¹, Rhonda Lightle¹, Thomas Moore¹, Ying Li¹, Dehua Bi^1,3, Timothy Carroll⁴, Yuan Ji³, Romuald Girard¹, and Issam A. Awad^1
1Neurovascular Surgery Program, Section of Neurosurgery, Department of Surgery, University of Chicago Medicine and Biological Sciences, Chicago, IL, United States, ²Bioinformatics Core, Center for Research Informatics, The University of Chicago, Chicago, IL, United States, ³Department of Public Health Sciences, The University of Chicago, Chicago, IL, United States, ⁴Department of Diagnostic Radiology, University of Chicago Medicine and Biological Sciences, Chicago, IL, United States

A cavernous angioma with symptomatic hemorrhage (CASH) is more likely to rebleed for several years while conventional MRI signatures of hemorrhage may disappear after a few weeks.  We aimed to investigate whether perfusion or permeability derivations of dynamic contrast-enhanced quantitative perfusion-MRI (DCEQP) can distinguish a lesion that had bled earlier or predict subsequent lesional bleeding/growth after DCEQP. Machine learning and Bayesian model selection showed that perfusion imaging may distinguish cases with CASH 3–12 months prior to the scan (diagnostic biomarker) while a combination of permeability and perfusion derivations may predict bleeding/growth in the subsequent year (prognostic biomarker).

Dandan Yang^1,2, Yongjun Han³, Dongye Li⁴, Huiyu Qiao², Hualu Han², Rui Shen², Zihan Ning², and Xihai Zhao^2
1Capital Medical University, Beijing, China, ²Tsinghua University, Beijing, China, ³Aerospace Center Hospital, Beijing, China, ⁴Sun Yat-Sen Memorial Hospital, Guangzhou, China

This study aimed to determine the distribution patterns of acute cerebral infarction (ACI) between young and elderly patients with intracranial large artery stenosis using diffusion-weighted imaging. In total, 69 young patients and 77 elderly patients were included in this study. We found that young patients had significantly higher prevalence of multiple ACI lesions in more than one vascular territory and higher percentage of lesions in both anterior and posterior circulations than elderly patients. Our findings provide new insights in optimizing the etiological diagnosis of cerebrovascular disease for young and elderly patients.

Mohammed A. Al-masni¹, Woo-Ram Kim², Eung Yeop Kim³, Young Noh⁴, and Dong-Hyun Kim^1
1Department of Electrical and Electronic Engineering, College of Engineering, Yonsei University, Seoul, Korea, Republic of, ²Neuroscience Research Institute, Gachon University, Incheon, Korea, Republic of, ³Department of Radiology, Gachon University College of Medicine, Gachon University, Incheon, Korea, Republic of, ⁴Department of Neurology, Gachon University College of Medicine, Gachon University, Incheon, Korea, Republic of

Lacunes are small cerebrospinal fluid-filled lesions that are generated by the occlusion of penetrating deep branches of cerebral arteries. Early detection of lacunes could decrease the possible clinical implications such as dementia, gait impairment, and lacunar stroke. In this study, we propose a deep learning 3D multi-scale residual network for lacunes identification using FLAIR and T1-MPRAGE MR images. We redesign the proposed network via applying multiple parallel paths using different input scales. This enables to extract more robust contextual global features and hence achieve better detection performance. The proposed work exhibits its ability to distinguish true lacunes from non-lacunes.

Su Yan¹, Guiling Zhang¹, Weiyin Vivian Liu², and Wenzhen Zhu^1
1Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, ²MR Research, GE Healthcare, Beijing, China

The primary aim of this study was to examine the changes of cortical morphology in patients with unilateral basal ganglia stroke (BS) in subacute phase and evaluate the discrepancies of structural covariance networks (SCNs) between BS and HCs groups by using a seed-based structural covariance approach. The main findings were that BS could cause cortical atrophy of bilateral frontal and temporal lobes and abnormal structural covariance patterns, featured by decreased global efficiency and fragile topological properties in reaction to target attacks. These findings may enable us to better understand the neurobiological mechanisms of behavioral impairment and recovery after BS.

Sivakami Avadiappan¹, Jeffrey Nelson², Marc Mabray³, Blaine Hart³, Leslie Morrison⁴, Atif Zafar⁴, Michel Torbey⁴, Helen Kim^2,5, and Janine Lupo^1
1Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, United States, ²Department of Anesthesia and Perioperative Care, University of California San Francisco, San Francisco, CA, United States, ³Department of Radiology, University of New Mexico, Albuquerque, NM, United States, ⁴Department of Neurology, University of New Mexico, Albuquerque, NM, United States, ⁵Department of Epidemiology and Biostatistics, University of California San Francisco, San Francisco, CA, United States

Cerebral cavernous malformations (CCMs) are vascular lesions composed of abnormally enlarged small blood vessels and are characterized by symptoms such as seizures and hemorrhages. In this study, we explore the relationship between serial MR imaging metrics and clinical markers of CCM disease severity. A semi-automated algorithm was applied on 57 patients to segment lesions. Age was associated with lesion count and obesity was inversely related to the lesion count. The lesion burdens in brainstem and temporal lobe were related to hemorrhagic and seizure events. These imaging metrics can function as surrogate markers for risk stratification in these patients. 

Mehdi Zoghinia¹, Mohammed Ayoub Alaoui Mhamdi¹, and Russell Butler^1,2
1Bishops university, sherbrooke, QC, Canada, ²Diagnostic Radiology, University of Sherbrooke, sherbrooke, QC, Canada

Paramagnetic deoxyhemoglobin introduces magnetic susceptibility to venous vessels in the brain, causing intra-voxel dephasing of surrounding tissue, appearing dark on a susceptibility-weighted image (SWI). However, empirical studies quantifying the extent of this dephasing is lacking. Using time-of-flight (TOF) angiogram as ground truth, and focusing on the straight sinus, we quantified the extent to which SWI vessel diameters deviated from true (TOF) vessel diameters. We find straight sinus diameter based on SWI segmentation to be nearly twice the diameter of TOF segmentation, and the difference between SWI and TOF straight sinus diameter was dependent on vessel location in the y-axis.

Chun-Jung Juan¹, Yi-Jui Liu², Shao-Chieh Lin³, and Yi-Hung Jeng^4
1Department of Medical Imaging, China Medical University Hsinchu Hospital, Hsinchu, Taiwan, ²Department of Automatic Control Engineering, Feng Chia University, Taichung, Taiwan, ³Ph.D. program in Electrical and Communication Engineering, Feng Chia University, Taichung, Taiwan, ⁴Department of Biomedical Engineering and Environmental Sciences, National Tsing Hua University, Hsinchu, Taiwan, Hsinchu, Taiwan

Accurate automatic segmentation of acute ischemic infarction on diffusion-weighted images (DWI) is clinically important. However, the accuracy of automatic segmentation of stroke lesions is affected by a lot of factors. By applying graded ADC thresholds, our study verifies the value of ADC threshold on the performance of the deep learning models in segmenting acute ischemic infarction with increasing the Dice similarity coefficient (DSC) by the lowering the ADC threshold. In addition, our study provides a new window to distinguish cytotoxic edema and vasogenic edema in acute stroke. Moreover, our results further show a size-dependent influence of DSC for stroke segmentation.

Shokufeh Sadaghiani¹, M. Dylan Tisdall², Sandhitsu R. Das¹, David A. Wolk¹, and John A. Detre^1
1Department of Neurology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, United States, ²Department of Radiology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, United States

We investigated the correlation of CMIs with MTL subregional cortical thickness in 21 healthy elderly subjects. We also showed that only 26% of total CMIs visible on 7T MRI scans are also visible on 3T scans suggesting a better visualization of CMIs by 7T MRI compared to 3T MRI. Our findings support an interaction between cerebrovascular and neurodegenerative mechanisms in healthy aging.

Huimin Mao¹, Weiqiang Dou², Xinyi Wang¹, Xinyu Wang¹, Kunjian Chen¹, and Yu Guo^1
1The First Affiliated Hospital of Shandong First Medical University, Jinan, China, ²GE Healthcare, MR Research China, Beijing, P.R. China, Beijing, China

The purpose of this study was to evaluate the imaging characteristics of intracranial large and small vessel diseases long after cranial irradiation. All patients with small vessel diseases underwent intracranial vessel wall imaging by 3D high resolution CUBE T1-weighted imaging (HR-MRI) to detect large vessel diseases. All recruited patients were found with intracranial small and large vessel diseases. Therefore, intracranial vasculopathy is not a rare complication after cranial irradiation, even in young patients. Patients after cranial irradiation should be followed up with MR imaging including HR-MRI.

Grant Nikseresht¹, Ashish A. Tamhane², Nazanin Makkinejad³, Carles Javierre-Petit³, Gady Agam¹, David A. Bennett², Julie A. Schneider², and Konstantinos Arfanakis^2,3
1Department of Computer Science, Illinois Institute of Technology, Chicago, IL, United States, ²Rush Alzheimer's Disease Center, Rush University Medical Center, Chicago, IL, United States, ³Department of Biomedical Engineering, Illinois Institute of Technology, Chicago, IL, United States

Cerebral microbleeds (CMBs) are common in older adults and have been linked to hypertension, increased risk of stroke, and cognitive decline. There is also evidence of an association between CMBs and cerebral amyloid angiopathy (CAA) in studies involving clinical populations and patients with intracerebral hemorrhage. However, it remains unclear what the neuropathologic correlates of CMBs are in community-based older adults. The aim of this study was to determine the neuropathologic correlates of CMBs in community-based older adults, and also to investigate the relationship between neuropathologies and the location of CMBs in the brain.

Dongsuk Sung¹, Peter A. Kottke², Jason W. Allen^1,3,4, Benjamin B. Risk⁵, Fadi Nahab⁴, Andrei G. Fedorov^2,6, and Candace C. Fleischer^1,3,6
1Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, GA, United States, ²Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA, United States, ³Department of Radiology and Imaging Sciences, Emory University School of Medicine, Atlanta, GA, United States, ⁴Department of Neurology, Emory University School of Medicine, Atlanta, GA, United States, ⁵Department of Biostatistics and Bioinformatics, Emory University, Atlanta, GA, United States, ⁶Petit Institute for Bioengineering and Bioscience, Georgia Institute of Technology, Atlanta, GA, United States

To prevent recurrent ischemic stroke, at-risk but viable tissue must be identified and reperfused. Conventional methods for identifying these regions can overestimate the infarct core leaving salvageable tissue untreated. To explore the use of brain temperature to identify viable tissue after stroke, we have refined our previous brain temperature model to simulate major cerebral artery occlusion and enable predictions of brain temperature evolution after stroke onset. Brain temperature was highest in ischemic penumbra followed by infarct core and lowest in healthy tissue, demonstrating its potential as an imaging biomarker for reperfusion therapy.

Ryan Andrew Rava^1,2, Kenneth V. Snyder^2,3, Muhammad Waqas^2,3, Elad I. Levy^2,3, Jason M. Davies^2,3, Adnan H. Siddiqui^2,3, and Ciprian N. Ionita^1,2,3
1Biomedical Engineering, University at Buffalo, Buffalo, NY, United States, ²Canon Stroke and Vascular Research Center, Buffalo, NY, United States, ³Neurosurgery, University at Buffalo, Buffalo, NY, United States

Convolutional neural networks have the potential to predict penumbra volumes within acute ischemic stroke patients to determine their eligibility for mechanical thrombectomy based on the Defuse 3 clinical trial. Currently, computed tomography perfusion is the main method used to quantify penumbra volumes but not all stroke centers have this modality available. In this study, 2 networks were developed to automatically segment penumbra using FLAIR and DWI and performance metrics comparing each network’s predictions with ground truth penumbra (dual network: Dice=0.61, sensitivity=0.68, PPV=0.59, multi-input network: Dice=0.61, sensitivity=0.62, PPV=0.64) indicate a multi-input network is the most capable of segmenting penumbra tissue.   

mengting wei¹, jinhao lv², liuxian wang², senhao zhang², dongshan han², xinrui wang², and xin lou^2
1Chinese PLA General Hospital, BeiJing, China, ²Chinese PLA General Hospital, beijing, China

Stroke is characterized by a high recurrence rate, and intervention after early identification of patients at risk of recurrence may improve their prognosis. After strict screening, 55 patients were enrolled in this study. the results show that it is feasible to identify patients with recurrent cerebrovascular events within one year by integrating clinical and imaging features through machine learning. RandomForest and NaiveBayes are the optimal algorithms, and HCR（hypoperfusion cubage ratio）can significantly optimize the recognition of these patients.

Jinhao Lyu¹, Mengting Wei¹, Xiangbing Bian¹, Liuxian Wang¹, Senhao Zhang¹, Lin Ma¹, and Xin Lou^1
1Radiology, Chinese PLA General Hospital, The First Medical Center, Beijing, China

Intracranial artery steno-occlusive disease causes high morbidity of ischemic stroke occurrence and reoccurrence. The present study tested the presence of Fluid-Attenuated Inversion Recovery vascular hyperintensity (FVH) to discriminate symptomatic patients from asymptomatic patients in intracranial internal carotid artery or middle cerebral artery occlusion and found that the presence of FVH was independently associated with a recent ischemic event in these patients and negatively correlated with collateral circulation. The findings suggest that FVH may serve as a feasible imaging marker to identify high-risk cases in their follow-up and clinical management.

Jing Yang¹, Yanwei Miao¹, Yangyingqiu Liu ¹, Yu Bing¹, Bingbing Gao¹, Jiazheng Wang², Zhiwei Shen², Ailian Liu¹, Qingwei Song¹, and Renwang Pu^1
1First Affiliated Hospital of Dalian Medical University, Dalian, China, ²Philips Healthcare, Dalian, China

The MR scan time can be decreased using Compressed sensing with a favorable image quality. The study aimed to quantitative measure the oxygen concentration using Susceptibility Weighted Imaging (SWI) images with the different acceleration factors (AF), and explore the optimized AF of compressed sensing based on quantitative assessment of the image quality. Combined with the SNR and CNR, when the AF is 4, the scanning time can be reduced by 71.39%. For the phase value of deep veins, the AF can reach to 6 and the scan time is reduced to 1min8s without image quality change obviously.

Liu-xian Wang¹, Dong-shan Han¹, Jin-hao Lyu¹, and Xin Lou^1
1the First Medical Center, Chinese PLA General Hospital, Beijing, China

Total SVD burden were not correlated with collateral perfusion, but perivascular space could impact collaterals. 

Catherine A. de Planque¹, Henk J. Mutsaerts², Vera C.W. Keil², Nicole S. Erler¹, Marjolein Dremmen¹, Irene M.J. Mathijssen¹, and Jan Peter^3
1Erasmus University Medical Center, Rotterdam, Netherlands, ²Amsterdam UMC, Amsterdam, Netherlands, ³Institute of Radiopharmaceutical Cancer Research, Dresden, Germany

Spatial normalization is an important step for image processing and quantification of regional brain perfusion values using arterial spin labeling (ASL) MRI and is typically performed via high-resolution structural scans. Structural segmentation and/or registration is complicated when gray-white matter T1w contrast is low and changing in early phases of myelination in newborns. Craniosynostosis is a condition where the decision for surgical treatment in the first years of life is supported by brain imaging. In this study, we investigate if ASL CBF image contrast can be directly used for spatial normalization, in both healthy controls and a non-syndromic type of craniosynostosis.

Eva van Grinsven¹, Anouk Smits¹, Emma van Kessel¹, Mathijs Raemaekers¹, Edward de Haan², Irene Huenges Wajer^1,3, Veerle Ruijters¹, Marielle Philippens⁴, Joost Verhoeff⁴, Pierre Robe¹, Tom Snijders¹, and Martine van Zandvoort^1,3
1Department of Neurology & Neurosurgery, University Medical Center Utrecht Brain Center, Utrecht University, Utrecht, Netherlands, ²Department of Psychology, University of Amsterdam, Amsterdam, Netherlands, ³Department of Experimental Psychology and Helmholtz Institute, Utrecht University, Utrecht, Netherlands, ⁴Department of Radiotherapy, University Medical Center Utrecht, Utrecht, Netherlands

While lesion-symptom mapping can inform on which brain regions are crucial for a given behavior, it is still unclear whether different lesion etiologies show comparable structure-function relationships. In this study, support-vector regression lesion-symptom maps were compared between a glioma and stroke population. As expected, pathology distinct coverage patterns in the brain were found and there were more and larger significant voxel clusters in the tumor group. Our preliminary conclusion is that despite some differences in lesion-symptom associations in comparing a tumor and stroke population, these two populations can provide complementary information regarding involvement of brain regions for given cognitive tasks.

Bejoy Thomas¹, Josline Elsa Joseph¹, Sabarish S Sekar¹, Santhosh Kannath¹, and Ramshekhar N Menon^2
1Imaging Sciences and Interventional Radiology, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Thiruvananthapuram, India, ²Neurology, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Thiruvananthapuram, India

Dural Arteriovenous fistula (dAVF) causes cognitive deficits and associated memory impairment. The objective of this study was to understand the frequency dependent correlates of resting state functional connectivity of memory deficit presentations in dAVF.  Our results reveal reduced functional connectivity, in a frequency dependent fashion, among hippocampus, parahippocampal cortex and dorsolateral prefrontal cortex, i.e.: regions involved in declarative memory functional processing.

Takashi Abe^1,2, Maki Otomo², Rintaro Ito¹, Rei Nakamichi¹, Yumi Abe¹, Toshiki Nakane¹, Hisashi Kawai¹, Toshiaki Taoka¹, Shinji Naganawa¹, and Masafumi Harada^2
1Nagoya University, Nagoya, Japan, ²Tokushima University, Tokushima, Japan

Although it is often difficult to differentiate brain abscess, acute ischemic stroke, and hemorrhagic stroke on diffusion weighted image (DWI), the diagnosis can be made more accurately by focusing on the low signal intensity, especially the low intensity rim surrounding the lesion. In this study, image interpretation of six diagnostic radiologists (12.8 years of experience on average) showed an accuracy of 70% before explanation of the low intensity rim finding, but the accuracy increased to 86% after explanation (p = 0.025), confirming the importance of focusing on low intensity rim on DWI.