scholarly journals BRAVE-NET: Fully Automated Arterial Brain Vessel Segmentation In Patients with Cerebrovascular Disease

Author(s):  
Adam Hilbert ◽  
Vince I. Madai ◽  
Ela M. Akay ◽  
Orhun U. Aydin ◽  
Jonas Behland ◽  
...  

AbstractIntroductionArterial brain vessel assessment is crucial for the diagnostic process in patients with cerebrovascular disease. Noninvasive neuroimaging techniques such as time-of-flight (TOF) magnetic resonance angiography (MRA) imaging are applied in the clinical routine to depict arteries. They are, however, only visually assessed. Fully automated vessel segmentation integrated into the clinical routine could facilitate the time-critical diagnosis of vessel abnormalities and might facilitate the identification of valuable biomarkers for cerebrovascular events. In the present work, we developed and validated a new deep learning model for vessel segmentation, coined BRAVE-NET, on a large aggregated dataset of patients with cerebrovascular diseases.MethodsBRAVE-NET is a multiscale 3-D convolutional neural network (CNN) model developed on a dataset of 264 patients from 3 different studies enrolling patients with cerebrovascular diseases. A context path, dually capturing high- and low-resolution volumes, and deep supervision were implemented. The BRAVE-NET model was compared to a baseline Unet model and variants with only context paths and deep supervision, respectively. The models were developed and validated using high-quality manual labels as ground truth. Next to precision and recall, the performance was assessed quantitatively by Dice coefficient (DSC); average Hausdorff distance (AVD); 95- percentile Hausdorff distance (95HD) and via visual qualitative rating.ResultsThe BRAVE-NET performance surpassed the other models for arterial brain vessel segmentation with a DSC = 0.931, AVD = 0.165 and 95HD = 29.153. The BRAVE-NET model was also the most resistant towards false labelings as revealed by the visual analysis. The performance improvement is primarily attributed to the integration of the multiscaling context path into the 3-D Unet and to a lesser extent to the deep supervision architectural component.DiscussionWe present a new state-of-the-art of arterial brain vessel segmentation tailored to cerebrovascular pathology. We provide an extensive experimental validation of the model using a large aggregated dataset encompassing a large variability of cerebrovascular disease. The framework provides the technological foundation for improving the clinical workflow and can serve as a biomarker extraction tool in cerebrovascular diseases.

2020 ◽  
Vol 3 ◽  
Author(s):  
Adam Hilbert ◽  
Vince I. Madai ◽  
Ela M. Akay ◽  
Orhun U. Aydin ◽  
Jonas Behland ◽  
...  

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Orhun Utku Aydin ◽  
Abdel Aziz Taha ◽  
Adam Hilbert ◽  
Ahmed A. Khalil ◽  
Ivana Galinovic ◽  
...  

Abstract Background Arterial brain vessel segmentation allows utilising clinically relevant information contained within the cerebral vascular tree. Currently, however, no standardised performance measure is available to evaluate the quality of cerebral vessel segmentations. Thus, we developed a performance measure selection framework based on manual visual scoring of simulated segmentation variations to find the most suitable measure for cerebral vessel segmentation. Methods To simulate segmentation variations, we manually created non-overlapping segmentation errors common in magnetic resonance angiography cerebral vessel segmentation. In 10 patients, we generated a set of approximately 300 simulated segmentation variations for each ground truth image. Each segmentation was visually scored based on a predefined scoring system and segmentations were ranked based on 22 performance measures common in the literature. The correlation of visual scores with performance measure rankings was calculated using the Spearman correlation coefficient. Results The distance-based performance measures balanced average Hausdorff distance (rank = 1) and average Hausdorff distance (rank = 2) provided the segmentation rankings with the highest average correlation with manual rankings. They were followed by overlap-based measures such as Dice coefficient (rank = 7), a standard performance measure in medical image segmentation. Conclusions Average Hausdorff distance-based measures should be used as a standard performance measure in evaluating cerebral vessel segmentation quality. They can identify more relevant segmentation errors, especially in high-quality segmentations. Our findings have the potential to accelerate the validation and development of novel vessel segmentation approaches.


1995 ◽  
Vol 73 (04) ◽  
pp. 558-560 ◽  
Author(s):  
Kimmo Kontula ◽  
Antti Ylikorkala ◽  
Helena Miettinen ◽  
Alpo Vuorio ◽  
Ritva Kauppinen-Mäkelin ◽  
...  

SummaryThe point mutation Arg506->Gln of factor V was recently shown to be an important and relatively common genetic cause of venous thromboembolism. Using a DNA technique based on polymerase chain reaction, we surveyed the blood samples of 236 patients with ischaemic stroke or a transient ischaemic attack, 122 survivors of myocardial infarction and 137 control subjects for the presence of this mutation. Although the frequency of the factor V mutation in patients with arterial disease (4.5%) was not significantly different from that in healthy blood donors (2.9%), a carrier status for this mutant gene was associated with symptoms of migraine and relatively mild angiographic abnormalities among patients with cerebrovascular disease. A more extensive study addressing the occurrence and significance of the mutant factor V mutation in patients with vasospastic cerebrovascular diseases seems to be warranted.


Author(s):  
Chao Zhang ◽  
Zhenyu Quan ◽  
Qincheng Wu ◽  
Zhezhen Jin ◽  
Joseph Lee ◽  
...  

Background: Air pollution in large Chinese cities has led to recent studies that highlighted the relationship between particulate matters (PM) and elevated risk of cardio-cerebrovascular mortality. However, it is unclear as to whether: (1) The same adverse relations exist in cities with relatively low levels of air pollution; and (2) the relationship between the two are similar across ethnic groups. Methods: We collected data of PM2.5 (PM with an aerodynamic diameter ≤ 2.5 µm) and PM10 (aerodynamic diameter ≤ 10 µm) in the Yanbian Korean Autonomous Prefecture between 1 January 2015 and 31 December 2016. Using a time-stratified case-crossover design, we investigated whether levels of particulate pollutants influence the risk of cardio-cerebrovascular disease mortality among ethnic Korean vs. ethnic Han residents residing in the Yanbian Korean Autonomous Prefecture. Results: Under the single air pollutant model, the odds ratios (ORs) of cardio-cerebrovascular disease were 1.025 (1.024–1.026) for each 10 μg/m3 increase in PM2.5 at lag0 day, 1.012 (1.011–1.013) for each 10 μg/m3 increase in PM10 at lag1 day. In the multi-pollutant model adjusted by PM10, SO2, and NO2, the ORs of cardio-cerebrovascular disease were 1.150 (1.145–1.155) for ethnic Koreans and 1.154 (1.149–1.158) for ethnic Hans for each 10 μg/m3 increase in PM2.5. In the multi-pollutant model adjusted by PM2.5, SO2, and NO2, the ORs of cardio-cerebrovascular disease were 1.050 (1.047–1.053) for ethnic Koreans and 1.041 (1.039–1.043) for ethnic Hans for each 10 μg/m3 increase in PM10. Conclusion: This study showed that PM2.5 and PM10 were associated with increased risks of acute death events in residential cardio-cerebrovascular disease in Yanbian, China.


2020 ◽  
Vol 31 (4) ◽  
pp. 401-410
Author(s):  
Carl Christofer Juhlin ◽  
Jan Zedenius ◽  
Anders Höög

AbstractNeuroendocrine neoplasms (NENs) have traditionally been identified via expression of proteins associated to the regulation of secretory vesicles and granules. We report the clinical usage of the “second-generation” proteins ISL LIM homeobox 1 (ISL1), INSM transcriptional repressor 1 (INSM1), and secretagogin (SECG) as immunohistochemical markers of neuroendocrine differentiation since their introduction in clinical routine and compare the results with the established proteins chromogranin A (CGA) and synaptophysin (SYP). In total, 161 tumors, including 139 NENs and 22 “non-NENs” (unrelated tumors with an initial suspicion of NEN), were informatively stained for ISL1, and subsets were also interrogated for INSM1 and/or SECG. Diffuse or focal positive immunoreactivity was noted for ISL1 in 91/139 NENs (65%) and in 6/22 (27%) non-NENs, for INSM1 in 76/85 NENs (89%) and in 2/5 (40%) non-NENs, and for SECG in 49 out of 64 NENs (77%) and in 0/5 non-NENs (0%). Generally, ISL1, INSM1, and SECG exhibited sensitivities in line with or slightly below that of CGA and SYP—largely attributable to tissue-specific patterns regarding tumoral origin. Moreover, for pancreatic and small intestinal NENs, the two largest subgroups, ISL1 staining results were consistent irrespectively of tumor source and WHO grade. We verify previously suggested immunohistochemical schemes of neuroendocrine markers of first- and second-generations to facilitate the diagnostic process for NENs and confirm that the second-generation neuroendocrine markers display tissue-specific patterns. We therefore recommend their implementation in tertiary endocrine pathology centers, not least to aid in the identification of primary tumors when analyzing metastases.


2018 ◽  
Vol 38 (9) ◽  
pp. 1387-1390 ◽  
Author(s):  
Manus J Donahue ◽  
Jeroen Hendrikse

Approximately 15 million individuals suffer a stroke worldwide each year, and stroke results in death or permanent disability in two-thirds of these individuals. Due to increased knowledge and management of modifiable risk factors, stroke incidence in developed countries is declining, however remains high at just under 1 million patients per year in the United States alone. Further improving management of patients with cerebrovascular disease (CVD) ultimately will require development and clinical adoption of sensitive markers of hemodynamic and metabolic failure, as well as trials that evaluate how to interpret these markers to optimize therapies. Realizing this goal and reducing the complete burden of CVD is dependent on an improved understanding of the pathophysiological processes that underlie CVD in all stages, including sub-clinical disease processes, acute stroke, and post-stroke recovery mechanisms. This document serves as an introduction to the Journal of Cerebral Blood Flow and Metabolism special issue on cerebrovascular diseases, which is comprised of contributions from experts in each of the above stages of CVD, and outlines current standards for patient management and emerging directions that have potential for improving patient care over the next decade.


2015 ◽  
Vol 39 (2) ◽  
pp. 138-143 ◽  
Author(s):  
Jacques L. De Reuck ◽  
Vincent Deramecourt ◽  
Florent Auger ◽  
Nicolas Durieux ◽  
Charlotte Cordonnier ◽  
...  

Background: As cortical microbleeds and microinfarcts in neurodegenerative and cerebrovascular diseases have been studied predominantly at the level of the cerebral hemispheres and linked to the presence of cerebral amyloid angiopathy (CAA), we aimed at determining with 7.0-tesla magnetic resonance imaging (MRI) whether the causes and the frequency of cortical cerebellar microbleeds (CCeMBs) and microinfarcts (CCeMIs) are the same. Materials and Methods: Hundred and four postmortem brains, composed of 29 with pure Alzheimer's disease (AD), 9 with AD associated to CAA, 10 with frontotemporal lobar degeneration, 9 with amyotrophic lateral sclerosis, 10 with Lewy body disease, 12 with progressive supranuclear palsy, 9 with vascular dementia (VaD), and 16 controls, were examined. On a horizontal section of a cerebellar hemisphere examined with 7.0-tesla MRI, the number CCeMBs and CCeMIs were compared between the different disease groups and the control group. The MRI findings were also compared with the corresponding mean values observed on histological examination of a separate standard horizontal section of a cerebellar hemisphere, used for diagnostic purpose. Results: CCeMBs and CCeMIs were only significantly increased in the VaD group. When comparing the diseased patients with and without CAA mutually and with those with arterial hypertension and severe atherosclerotic cerebrovascular disease, only in the latter an increase of CCeMBs and CCeMIs was observed. There was an excellent correlation between the MRI and the neuropathological findings. Conclusions: CCeMBs and CCeMIs are mainly due to atherosclerotic cerebrovascular disease and not due to CAA. Their increased presence cannot be included to the Boston diagnostic criteria for CAA.


1987 ◽  
Vol 15 (3) ◽  
pp. 153-157 ◽  
Author(s):  
Mats Brommels ◽  
Reijo Tilvis ◽  
Lauri Autio

A total of 20680 patients hospitalised in Helsinki during 1970–1980 due to cerebrovascular diseases were found when screening the Finnish National Hospital Discharge Register. The material was categorised according to three-digit ICD-8 diagnosis codes and age, and was analysed for case-fatality, length of stay and discharge status. By identifying all new cases an assessment of the incidence development during the study period was also possible. A fall in the overall age-standardised incidence of cerebrovascular disease was demonstrated, in accordance with disease register studies. The main reason was decline in incidence of haemorrhagic stoke (ICD-8 no. 431) and less well defined types of stroke (436–438). Ischaemic stroke (433), on the other hand, did not decrease in frequency. The diagnostic shift, occurring parallell with a growing mean age of patients, lead to decreasing acute mortality, increasing institutionalisation rates and longer stays in hospital, thus resulting in growing figures of hospital utilisation in spite of the declining incidence.


2020 ◽  
Vol 12 (5) ◽  
pp. 92-97
Author(s):  
L. M. Antonenko ◽  
N. V. Vakhnina ◽  
D. O. Gromova

Hypertension is a widespread disease related to modifiable vascular risk factors for stroke and chronic cerebrovascular diseases. The pathogenetic basis of brain damage in hypertension is cerebral microangiopathy that leads to vascular cognitive impairment (CI), instability, and falls. Microcirculatory changes in the presence of hypertension at the initial stages of cerebrovascular disease occur without visible clinical manifestations of brain damage. Pathogenetically justified treatment used at an early stage of the disease makes it possible to achieve good results in the prevention of vascular brain damage. An important aspect of selecting effective therapy is the competent diagnosis of the causes of dizziness and instability, which can be caused not only by brain damage, but also by peripheral vestibular system diseases. Early diagnosis of vascular CI, selection of adequate therapy, and prevention of their further progression are of great importance. The studies performed have shown the high efficacy of vinpocetine (Cavinton®) that has a multifactorial mechanism of action in the treatment and prevention of CI, dizziness, and instability caused by cerebrovascular disease.


Sign in / Sign up

Export Citation Format

Share Document