Brain vessel status is a promising biomarker for better prevention and treatment in cerebrovascular disease. However, classic rule-based vessel segmentation algorithms need to be hand-crafted and are ...insufficiently validated. A specialized deep learning method-the U-net-is a promising alternative. Using labeled data from 66 patients with cerebrovascular disease, the U-net framework was optimized and evaluated with three metrics: Dice coefficient, 95% Hausdorff distance (95HD) and average Hausdorff distance (AVD). The model performance was compared with the traditional segmentation method of graph-cuts. Training and reconstruction was performed using 2D patches. A full and a reduced architecture with less parameters were trained. We performed both quantitative and qualitative analyses. The U-net models yielded high performance for both the full and the reduced architecture: A Dice value of ~0.88, a 95HD of ~47 voxels and an AVD of ~0.4 voxels. The visual analysis revealed excellent performance in large vessels and sufficient performance in small vessels. Pathologies like cortical laminar necrosis and a rete mirabile led to limited segmentation performance in few patients. The U-net outperfomed the traditional graph-cuts method (Dice ~0.76, 95HD ~59, AVD ~1.97). Our work highly encourages the development of clinically applicable segmentation tools based on deep learning. Future works should focus on improved segmentation of small vessels and methodologies to deal with specific pathologies.
•We developed an artificial intelligence model that generates 5 perfusion parameter maps from time-of-flight magnetic resonance angiography using image-to-image translation.•The generated perfusion ...parameter maps showed high visual similarity and moderate quantitative overlap with the real perfusion maps and areas of hypoperfusion.•This approach could offer an alternative to contrast agent-based imaging.•Before clinical use of this method for patient stratification and assessment of cerebral hemodynamics in cerebrovascular disease patients, validation is warranted.
To generate perfusion parameter maps from Time-of-flight magnetic resonance angiography (TOF-MRA) images using artificial intelligence to provide an alternative to traditional perfusion imaging techniques.
This retrospective study included a total of 272 patients with cerebrovascular diseases; 200 with acute stroke (from 2010 to 2018), and 72 with steno-occlusive disease (from 2011 to 2014). For each patient the TOF MRA image and the corresponding Dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI) were retrieved from the datasets. The authors propose an adapted generative adversarial network (GAN) architecture, 3D pix2pix GAN, that generates common perfusion maps (CBF, CBV, MTT, TTP, Tmax) from TOF-MRA images. The performance was evaluated by the structural similarity index measure (SSIM). For a subset of 20 patients from the acute stroke dataset, the Dice coefficient was calculated to measure the overlap between the generated and real hypoperfused lesions with a time-to-maximum (Tmax) > 6 s.
The GAN model exhibited high visual overlap and performance for all perfusion maps in both datasets: acute stroke (mean SSIM 0.88–0.92, mean PSNR 28.48–30.89, mean MAE 0.02–0.04 and mean NRMSE 0.14–0.37) and steno-occlusive disease patients (mean SSIM 0.83–0.98, mean PSNR 23.62–38.21, mean MAE 0.01–0.05 and mean NRMSE 0.03–0.15). For the overlap analysis for lesions with Tmax>6 s, the median Dice coefficient was 0.49.
Our AI model can successfully generate perfusion parameter maps from TOF-MRA images, paving the way for a non-invasive alternative for assessing cerebral hemodynamics in cerebrovascular disease patients. This method could impact the stratification of patients with cerebrovascular diseases. Our results warrant more extensive refinement and validation of the method.
Objective
Early identification of patients at risk of space‐occupying “malignant” middle cerebral artery (MCA) infarction (MMI) is needed to enable timely decision for potentially life‐saving ...treatment such as decompressive hemicraniectomy. We tested the hypothesis that acute stroke magnetic resonance imaging (MRI) predicts MMI within 6 hours of stroke onset.
Methods
In a prospective, multicenter, observational cohort study patients with acute ischemic stroke and MCA main stem occlusion were studied by MRI including diffusion‐weighted imaging (DWI), perfusion imaging (PI), and MR‐angiography within 6 hours of symptom onset. Multivariate regression analysis was used to identify clinical and imaging predictors of MMI.
Results
Of 140 patients included, 27 (19.3%) developed MMI. The following parameters were identified as independent predictors of MMI: larger acute DWI lesion volume (per 1 ml odds ratio OR 1.04, 95% confidence interval CI 1.02–1.06; p < 0.001), combined MCA + internal carotid artery occlusion (5.38, 1.55–18.68; p = 0.008), and severity of neurological deficit on admission assessed by the National Institutes of Health Stroke Scale score (per 1 point 1.16, 1.00–1.35; p = 0.053). The prespecified threshold of a DWI lesion volume >82 ml predicted MMI with high specificity (0.98, 95% CI 0.94–1.00), negative predictive value (0.90, 0.83–0.94), and positive predictive value (0.88, 0.62–0.98), but sensitivity was low (0.52, 0.32–0.71).
Interpretation
Stroke MRI on admission predicts malignant course in severe MCA stroke with high positive and negative predictive value and may help in guiding treatment decisions, such as decompressive surgery. In a subset of patients with small initial DWI lesion volumes, repeated diagnostic tests are required. ANN NEUROL 2010
Arterial brain vessel assessment is crucial for the diagnostic process in patients with cerebrovascular disease. Non-invasive 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.
BRAVE-NET is a multiscale 3-D convolutional neural network (CNN) model developed on a dataset of 264 patients from three 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.
The 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 toward 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.
We 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 and an external set of healthy volunteers. The framework provides the technological foundation for improving the clinical workflow and can serve as a biomarker extraction tool in cerebrovascular diseases.
In ischemic stroke, positron-emission tomography (PET) established the imaging-based concept of penumbra. It defines hypoperfused, but functionally impaired, tissue with preserved viability that can ...be rescued by timely reperfusion. Diffusion-weighted and perfusion-weighted (PW) magnetic resonance imaging (MRI) translated the concept of penumbra to the concept of mismatch. However, the use of mismatch-based patient stratification for reperfusion therapy remains a matter of debate. The equivalence of mismatch and penumbra, as well as the validity of the classical mismatch concept is questioned for several reasons. First, methodological differences between PET and MRI lead to different definitions of the tissue at risk. Second, the mismatch concept is still poorly standardized among imaging facilities causing relevant variability in stroke research. Third, relevant conceptual issues (e.g., the choice of the adequate perfusion measure, the best quantitative approach to perfusion maps, and the required size of the mismatch) need further refinement. Fourth, the use of single thresholds does not account for the physiological heterogeneity of the penumbra and probabilistic approaches may be more promising. The implementation of this current knowledge into an optimized state-of-the-art mismatch model and its validation in clinical stroke studies remains a major challenge for future stroke research.
Recent innovations such as CT installation in ambulances may lead to earlier start of stroke-specific treatments. However, such technically complex mobile facilities require effective methods of ...correctly identifying patients before deployment. We aimed to develop and validate a new dispatcher identification algorithm for stroke emergencies.
Dispatcher identification algorithm for stroke emergencies was informed by systematic qualitative analysis of the content of emergency calls to ambulance dispatchers for patients with stroke or transient ischemic attack (N=117) and other neurological (N=39) and nonneurological (N=51) diseases (Part A). After training of dispatchers, sensitivity and predictive values were determined prospectively in patients admitted to Charité hospitals by using the discharge diagnosis as reference standard (Part B).
Part A: Dysphasic/dysarthric symptoms (33%), unilateral symptoms (22%) and explicitly stated suspicion of stroke (47%) were typically identified in patients with stroke but infrequently in nonstroke cases (all <10%). Convulsive symptoms (41%) were frequent in other neurological diseases but not strokes (3%). Pain (26%) and breathlessness (31%) were often expressed in nonneurological emergencies (6% and 7% in strokes). Part B: Between October 15 and December 16, 2010, 5774 patients were admitted by ambulance with 246 coded with final stroke diagnoses. Sensitivity of dispatcher identification algorithm for stroke emergencies for detecting stroke was 53.3% and positive predictive value was 47.8% for stroke and 59.1% for stroke and transient ischemic attack. Of all 275 patients with stroke dispatcher codes, 215 (78.5%) were confirmed with neurological diagnosis.
Using dispatcher identification algorithm for stroke emergencies, more than half of all patients with stroke admitted by ambulance were correctly identified by dispatchers. Most false-positive stroke codes had other neurological diagnoses.
Intracranial atherosclerotic disease (ICAD) poses a significant risk of subsequent stroke but current prevention strategies are limited. Mechanistic simulations of brain hemodynamics offer an ...alternative precision medicine approach by utilising individual patient characteristics. For clinical use, however, current simulation frameworks have insufficient validation. In this study, we performed the first quantitative validation of a simulation-based precision medicine framework to assess cerebral hemodynamics in patients with ICAD against clinical standard perfusion imaging. In a retrospective analysis, we used a 0-dimensional simulation model to detect brain areas that are hemodynamically vulnerable to subsequent stroke. The main outcome measures were sensitivity, specificity, and area under the receiver operating characteristics curve (ROC AUC) of the simulation to identify brain areas vulnerable to subsequent stroke as defined by quantitative measurements of relative mean transit time (relMTT) from dynamic susceptibility contrast MRI (DSC-MRI). In 68 subjects with unilateral stenosis >70% of the internal carotid artery (ICA) or middle cerebral artery (MCA), the sensitivity and specificity of the simulation were 0.65 and 0.67, respectively. The ROC AUC was 0.68. The low-to-moderate accuracy of the simulation may be attributed to assumptions of Newtonian blood flow, rigid vessel walls, and the use of time-of-flight MRI for geometric representation of subject vasculature. Future simulation approaches should focus on integrating additional patient data, increasing accessibility of precision medicine tools to clinicians, addressing disease burden disparities amongst different populations, and quantifying patient benefit. Our results underscore the need for further improvement of mechanistic simulations of brain hemodynamics to foster the translation of the technology to clinical practice.
Stroke is a major cause of death or disability. As imaging-based patient stratification improves acute stroke therapy, dynamic susceptibility contrast magnetic resonance imaging (DSC-MRI) is of major ...interest in image brain perfusion. However, expert-level perfusion maps require a manual or semi-manual post-processing by a medical expert making the procedure time-consuming and less-standardized. Modern machine learning methods such as generative adversarial networks (GANs) have the potential to automate the perfusion map generation on an expert level without manual validation. We propose a modified pix2pix GAN with a temporal component (temp-pix2pix-GAN) that generates perfusion maps in an end-to-end fashion. We train our model on perfusion maps infused with expert knowledge to encode it into the GANs. The performance was trained and evaluated using the structural similarity index measure (SSIM) on two datasets including patients with acute stroke and the steno-occlusive disease. Our temp-pix2pix architecture showed high performance on the acute stroke dataset for all perfusion maps (mean SSIM 0.92-0.99) and good performance on data including patients with the steno-occlusive disease (mean SSIM 0.84-0.99). While clinical validation is still necessary for future studies, our results mark an important step toward automated expert-level perfusion maps and thus fast patient stratification.
Cerebrovascular disease, in particular stroke, is a major public health challenge. An important biomarker is cerebral hemodynamics. To measure and quantify cerebral hemodynamics, however, only ...invasive, potentially harmful or time-to-treatment prolonging methods are available.
We present a simulation-based approach which allows calculation of cerebral hemodynamics based on the patient-individual vessel configuration derived from structural vessel imaging. For this, we implemented a framework allowing segmentation and annotation of brain vessels from structural imaging followed by 0-dimensional lumped simulation modeling of cerebral hemodynamics. For annotation, a 3D-graphical user interface was implemented. For 0D-simulation, we used a modified nodal analysis, which was adapted for easy implementation by code. The simulation enables identification of areas vulnerable to stroke and simulation of changes due to different systemic blood pressures. Moreover, sensitivity analysis was implemented allowing the live simulation of changes to simulate procedures and disease progression. Beyond presentation of the framework, we demonstrated in an exploratory analysis in 67 patients that the simulation has a high specificity and low-to-moderate sensitivity to detect perfusion changes in classic perfusion imaging.
The presented precision medicine approach using novel biomarkers has the potential to make the application of harmful and complex perfusion methods obsolete.