Risk Factors for Perinatal Arterial Ischemic Stroke: A Machine Learning Approach.

BACKGROUND AND OBJECTIVES: Perinatal arterial ischemic stroke (PAIS) is a focal vascular brain injury presumed to occur between the fetal period and the first 28 days of life. It is the leading cause of hemiparetic cerebral palsy. Multiple maternal, intrapartum, delivery, and fetal factors have been associated with PAIS, but studies are limited by modest sample sizes and complex interactions between factors. Machine learning approaches use large and complex data sets to enable unbiased identification of clinical predictors but have not yet been applied to PAIS. We combined large PAIS data sets and used machine learning methods to identify clinical PAIS factors and compare this data-driven approach with previously described literature-driven clinical prediction models. METHODS: Common data elements from 3 registries with patients with PAIS, the Alberta Perinatal Stroke Project, Canadian Cerebral Palsy Registry, International Pediatric Stroke Study, and a longitudinal cohort of healthy controls (Alberta Pregnancy Outcomes and Nutrition Study), were used to identify potential predictors of PAIS. Inclusion criteria were term birth and idiopathic PAIS (absence of primary causative medical condition). Data including maternal/pregnancy, intrapartum, and neonatal factors were collected between January 2003 and March 2020. Common data elements were entered into a validated random forest machine learning pipeline to identify the highest predictive features and develop a predictive model. Univariable analyses were completed post hoc to assess the relationship between each predictor and outcome. RESULTS: A machine learning model was developed using data from 2,571 neonates, including 527 cases (20%) and 2,044 controls (80%). With a mean of 21 features selected, the random forest machine learning approach predicted the outcome with approximately 86.5% balanced accuracy. Factors that were selected a priori through literature-driven variable selection that were also identified as most important by the machine learning model were maternal age, recreational substance exposure, tobacco exposure, intrapartum maternal fever, and low Apgar score at 5 minutes. Additional variables identified through machine learning included in utero alcohol exposure, infertility, miscarriage, primigravida, meconium, spontaneous vaginal delivery, neonatal head circumference, and 1-minute Apgar score. Overall, the machine learning model performed better (area under the curve [AUC] 0.93) than the literature-driven model (AUC 0.73). DISCUSSION: Machine learning may be an alternative, unbiased method to identify clinical predictors associated with PAIS. Identification of previously suggested and novel clinical factors requires cautious interpretation but supports the multifactorial nature of PAIS pathophysiology. Our results suggest that identification of neonates at risk of PAIS is possible.

Spatial normalization for voxel-based lesion symptom mapping: impact of registration approaches.

Background: Voxel-based lesion symptom mapping (VLSM) assesses the relation of lesion location at a voxel level with a specific clinical or functional outcome measure at a population level. Spatial normalization, that is, mapping the patient images into an atlas coordinate system, is an essential pre-processing step of VLSM. However, no consensus exists on the optimal registration approach to compute the transformation nor are downstream effects on VLSM statistics explored. In this work, we evaluate four registration approaches commonly used in VLSM pipelines: affine (AR), nonlinear (NLR), nonlinear with cost function masking (CFM), and enantiomorphic registration (ENR). The evaluation is based on a standard VLSM scenario: the analysis of statistical relations of brain voxels and regions in imaging data acquired early after stroke onset with follow-up modified Rankin Scale (mRS) values. Materials and methods: Fluid-attenuated inversion recovery (FLAIR) MRI data from 122 acute ischemic stroke patients acquired between 2 and 3 days after stroke onset and corresponding lesion segmentations, and 30 days mRS values from a European multicenter stroke imaging study (I-KNOW) were available and used in this study. The relation of the voxel location with follow-up mRS was assessed by uni- as well as multi-variate statistical testing based on the lesion segmentations registered using the four different methods (AR, NLR, CFM, ENR; implementation based on the ANTs toolkit). Results: The brain areas evaluated as important for follow-up mRS were largely consistent across the registration approaches. However, NLR, CFM, and ENR led to distortions in the patient images after the corresponding nonlinear transformations were applied. In addition, local structures (for instance the lateral ventricles) and adjacent brain areas remained insufficiently aligned with corresponding atlas structures even after nonlinear registration. Conclusions: For VLSM study designs and imaging data similar to the present work, an additional benefit of nonlinear registration variants for spatial normalization seems questionable. Related distortions in the normalized images lead to uncertainties in the VLSM analyses and may offset the theoretical benefits of nonlinear registration.

Lesion-symptom mapping with NIHSS sub-scores in ischemic stroke patients.

BACKGROUND: Lesion-symptom mapping (LSM) is a statistical technique to investigate the population-specific relationship between structural integrity and post-stroke clinical outcome. In clinical practice, patients are commonly evaluated using the National Institutes of Health Stroke Scale (NIHSS), an 11-domain clinical score to quantitate neurological deficits due to stroke. So far, LSM studies have mostly used the total NIHSS score for analysis, which might not uncover subtle structure-function relationships associated with the specific sub-domains of the NIHSS evaluation. Thus, the aim of this work was to investigate the feasibility to perform LSM analyses with sub-score information to reveal category-specific structure-function relationships that a total score may not reveal. METHODS: Employing a multivariate technique, LSM analyses were conducted using a sample of 180 patients with NIHSS assessment at 48-hour post-stroke from the ESCAPE trial. The NIHSS domains were grouped into six categories using two schemes. LSM was conducted for each category of the two groupings and the total NIHSS score. RESULTS: Sub-score LSMs not only identify most of the brain regions that are identified as critical by the total NIHSS score but also reveal additional brain regions critical to each function category of the NIHSS assessment without requiring extensive, specialised assessments. CONCLUSION: These findings show that widely available sub-scores of clinical outcome assessments can be used to investigate more specific structure-function relationships, which may improve predictive modelling of stroke outcomes in the context of modern clinical stroke assessments and neuroimaging. TRIAL REGISTRATION NUMBER: NCT01778335.

Integrating expert knowledge for dementia risk prediction in individuals with mild cognitive impairment (MCI): a study protocol.

INTRODUCTION: To date, there is no broadly accepted dementia risk score for use in individuals with mild cognitive impairment (MCI), partly because there are few large datasets available for model development. When evidence is limited, the knowledge and experience of experts becomes more crucial for risk stratification and providing MCI patients with prognosis. Structured expert elicitation (SEE) includes formal methods to quantify experts' beliefs and help experts to express their beliefs in a quantitative form, reducing biases in the process. This study proposes to (1) assess experts' beliefs about important predictors for 3-year dementia risk in persons with MCI through SEE methodology and (2) to integrate expert knowledge and patient data to derive dementia risk scores in persons with MCI using a Bayesian approach. METHODS AND ANALYSIS: This study will use a combination of SEE methodology, prospectively collected clinical data, and statistical modelling to derive a dementia risk score in persons with MCI . Clinical expert knowledge will be quantified using SEE methodology that involves the selection and training of the experts, administration of questionnaire for eliciting expert knowledge, discussion meetings and results aggregation. Patient data from the Prospective Registry for Persons with Memory Symptoms of the Cognitive Neurosciences Clinic at the University of Calgary; the Alzheimer's Disease Neuroimaging Initiative; and the National Alzheimer's Coordinating Center's Uniform Data Set will be used for model training and validation. Bayesian Cox models will be used to incorporate patient data and elicited data to predict 3-year dementia risk. DISCUSSION: This study will develop a robust dementia risk score that incorporates clinician expert knowledge with patient data for accurate risk stratification, prognosis and management of dementia.

Correction: Improved multi-parametric prediction of tissue outcome in acute ischemic stroke patients using spatial features.

[This corrects the article DOI: 10.1371/journal.pone.0228113.].

Improved multi-parametric prediction of tissue outcome in acute ischemic stroke patients using spatial features.

INTRODUCTION: In recent years, numerous methods have been proposed to predict tissue outcome in acute stroke patients using machine learning methods incorporating multiparametric imaging data. Most methods include diffusion and perfusion parameters as image-based parameters but do not include any spatial information although these parameters are spatially dependent, e.g. different perfusion properties in white and gray brain matter. This study aims to investigate if including spatial features improves the accuracy of multi-parametric tissue outcome prediction. MATERIALS AND METHODS: Acute and follow-up multi-center MRI datasets of 99 patients were available for this study. Logistic regression, random forest, and XGBoost machine learning models were trained and tested using acute MR diffusion and perfusion features and known follow-up lesions. Different combinations of atlas coordinates and lesion probability maps were included as spatial information. The stroke lesion predictions were compared to the true tissue outcomes using the area under the receiver operating characteristic curve (ROC AUC) and the Dice metric. RESULTS: The statistical analysis revealed that including spatial features significantly improves the tissue outcome prediction. Overall, the XGBoost and random forest models performed best in every setting and achieved state-of-the-art results regarding both metrics with similar improvements achieved including Montreal Neurological Institute (MNI) reference space coordinates or voxel-wise lesion probabilities. CONCLUSION: Spatial features should be integrated to improve lesion outcome prediction using machine learning models.

Segmentation-Based Blood Flow Parameter Refinement in Cerebrovascular Structures Using 4-D Arterial Spin Labeling MRA.

OBJECTIVE: Cerebrovascular diseases are one of the main global causes of death and disability in the adult population. The preferred imaging modality for the diagnostic routine is digital subtraction angiography, an invasive modality. Time-resolved three-dimensional arterial spin labeling magnetic resonance angiography (4D ASL MRA) is an alternative non-invasive modality, which captures morphological and blood flow data of the cerebrovascular system, with high spatial and temporal resolution. This work proposes advanced medical image processing methods that extract the anatomical and hemodynamic information contained in 4D ASL MRA datasets. METHODS: A previously published segmentation method, which uses blood flow data to improve its accuracy, is extended to estimate blood flow parameters by fitting a mathematical model to the measured vascular signal. The estimated values are then refined using regression techniques within the cerebrovascular segmentation. The proposed method was evaluated using fifteen 4D ASL MRA phantoms, with ground-truth morphological and hemodynamic data, fifteen 4D ASL MRA datasets acquired from healthy volunteers, and two 4D ASL MRA datasets from patients with a stenosis. RESULTS: The proposed method reached an average Dice similarity coefficient of 0.957 and 0.938 in the phantom and real dataset segmentation evaluations, respectively. The estimated blood flow parameter values are more similar to the ground-truth values after the refinement step, when using phantoms. A qualitative analysis showed that the refined blood flow estimation is more realistic compared to the raw hemodynamic parameters. CONCLUSION: The proposed method can provide accurate segmentations and blood flow parameter estimations in the cerebrovascular system using 4D ASL MRA datasets. SIGNIFICANCE: The information obtained with the proposed method can help clinicians and researchers to study the cerebrovascular system non-invasively.

Automatic classification of major depression disorder using arterial spin labeling MRI perfusion measurements.

AIM: Neuroimaging-based multivariate pattern-recognition methods have been successfully used to develop diagnostic algorithms to distinguish patients with major depressive disorder (MDD) from healthy controls (HC). We developed and evaluated the accuracy of a multivariate classification method for the differentiation of MDD and HC using cerebral blood flow (CBF) features measured by non-invasive arterial spin labeling (ASL) MRI. METHODS: Twenty-two medication-free patients with the diagnosis of MDD based on DSM-IV criteria and 22 HC underwent pseudo-continuous 3-D-ASL imaging to assess CBF. Using an atlas-based approach, regional CBF was determined in various brain regions and used together with sex and age as classification features. A linear kernel support vector machine was used for feature ranking and selection as well as for the classification of patients with MDD and HC. Permutation testing was used to test for significance of the classification results. RESULTS: The automatic classifier based on CBF features showed a statistically significant accuracy of 77.3% (P = 0.004) with a specificity of 80% and sensitivity of 75% for classification of MDD versus HC. The features that contributed to the classification were sex and regional CBF of the cortical, limbic, and paralimbic regions. CONCLUSION: Machine-learning models based on CBF measurements are capable of differentiating MDD from HC with high accuracy. The use of larger study cohorts and inclusion of other imaging measures may improve the performance of the classifier to achieve the accuracy required for clinical application.

Hemodynamic Differences Between Recurrent and Nonrecurrent Intracranial Aneurysms: Fluid Dynamics Simulations Based on MR Angiography.

BACKGROUND AND PURPOSE: Although the role of wall shear stress (WSS) in the initiation, growth, and rupture of intracranial aneurysms has been well studied, its influence on aneurysm recurrence after endovascular treatment requires further investigation. We aimed to compare WSS at necks of recurrent and nonrecurrent aneurysms. METHODS: Nine recurrent coil-embolized aneurysms were identified and matched with nine nonrecurrent aneurysms. Patient-specific vessel geometries reconstructed from follow-up 3-D time-of-flight magnetic resonance angiography were analyzed using computational fluid dynamics (CFD) simulations. Absolute WSS and the percentage of abnormally low and high WSS at the aneurysm neck compared to the near artery were measured. RESULTS: The median percentage of abnormal WSS at the aneurysm neck was 49.3% for recurrent and 34.7% for nonrecurrent aneurysms (P = .011). The area under the receiver-operating-characteristic curve for distinguishing these aneurysms according to the percentage of abnormal WSS was .86 (95% CI .62 to .98). The optimal cut-off value of 45.1% resulted in a sensitivity and a specificity of 88.89% (95% CI 51.8% to 99.7%). CONCLUSION: Our findings indicate that necks of recurrent aneurysms are exposed to abnormal WSS to a larger extent. Abnormal WSS may serve as a metric to distinguish them from nonrecurrent aneurysms with CFD simulations a priori.

DCE MRI reveals early decreased and later increased placenta perfusion after a stress challenge during pregnancy in a mouse model.

OBJECTIVES: Stress during pregnancy is known to have negative effects on fetal outcome. The purpose of this exploratory study was to examine placental perfusion alterations after stress challenge during pregnancy in a mouse model. MATERIAL AND METHODS: Seven Tesla MRI was performed on pregnant mice at embrionic day (ED) 14.5 and 16.5. Twenty dams were exposed to an established acoustic stress challenge model while twenty non-exposed dams served as controls. Placental perfusion was analyzed in dynamic contrast-enhanced (DCE) MRI using the steepest slope model. The two functional placental compartments, the highly vascularized labyrinth and the endocrine junctional zone, were assessed separately. RESULTS: Statistical analysis revealed decreased perfusion levels in the stress group at ED 14.5 compared to controls in both placenta compartments. On ED 16.5, the perfusion level increased significantly in the stress group while placenta perfusion in controls remained similar or even slightly decreased leading to an overall increased perfusion in the stress group on ED 16.5 compared to controls. CONCLUSION: MR imaging allows noninvasive placenta perfusion assessment in this fetal stress mimicking animal model. In this exploratory study, we demonstrated that stress challenge during pregnancy leads to an initial reduction followed by an increase of placenta perfusion.

Automatic Temporal Segmentation of Vessels of the Brain Using 4D ASL MRA Images.

OBJECTIVE: Automatic vessel segmentation can be used to process the considerable amount of data generated by four-dimensional arterial spin labeling magnetic resonance angiography (4D ASL MRA) images. Previous segmentation approaches for dynamic series of images propose either reducing the series to a temporal average (tAIP) or maximum intensity projection (tMIP) prior to vessel segmentation, or a separate segmentation of each image. This paper introduces a method that combines both approaches to overcome the specific drawbacks of each technique. METHODS: Vessels in the tAIP are enhanced by using the ranking orientation responses of path operators and multiscale vesselness enhancement filters. Then, tAIP segmentation is performed using a seed-based algorithm. In parallel, this algorithm is also used to segment each frame of the series and identify small vessels, which might have been lost in the tAIP segmentation. The results of each individual time frame segmentation are fused using an or boolean operation. Finally, small vessels found only in the fused segmentation are added to the tAIP segmentation. RESULTS: In a quantitative analysis using ten 4D ASL MRA image series from healthy volunteers, the proposed combined approach reached an average Dice coefficient of 0.931, being more accurate than the corresponding tMIP, tAIP, and single time frame segmentation methods with statistical significance. CONCLUSION: The novel combined vessel segmentation strategy can be used to obtain improved vessel segmentation results from 4D ASL MRA and other dynamic series of images. SIGNIFICANCE: Improved vessel segmentation of 4D ASL MRA allows a fast and accurate assessment of cerebrovascular structures.

Stroke subtype classification by geometrical descriptors of lesion shape.

BACKGROUND AND PURPOSE: Inference of etiology from lesion pattern in acute magnetic resonance imaging is valuable for management and prognosis of acute stroke patients. This study aims to assess the value of three-dimensional geometrical lesion-shape descriptors for stroke-subtype classification, specifically regarding stroke of cardioembolic origin. METHODS: Stroke Etiology was classified according to ASCOD in retrospectively selected patients with acute stroke. Lesions were segmented on diffusion-weighed datasets, and descriptors of lesion shape quantified: surface area, sphericity, bounding box volume, and ratio between bounding box and lesion volume. Morphological measures were compared between stroke subtypes classified by ASCOD and between patients with embolic stroke of cardiac and non-cardiac source. RESULTS: 150 patients (mean age 77 years; 95% CI, 65-80 years; median NIHSS 6, range 0-22) were included. Group comparison of lesion shape measures demonstrated that lesions caused by small-vessel disease were smaller and more spherical compared to other stroke subtypes. No significant differences of morphological measures were detected between patients with cardioembolic and non-cardioembolic stroke. CONCLUSION: Stroke lesions caused by small vessel disease can be distinguished from other stroke lesions based on distinctive morphological properties. However, within the group of embolic strokes, etiology could not be inferred from the morphology measures studied in our analysis.

CT-perfusion stroke imaging: a threshold free probabilistic approach to predict infarct volume compared to traditional ischemic thresholds.

The aim was to evaluate a novel method of threshold-free prediction of brain infarct from computed tomography perfusion (CTP) imaging in comparison to conventional ischemic thresholds. In a multicenter cohort of 161 patients with acute large vessel occlusion who received endovascular therapy, brain infarction was predicted by CTP using (1) optimized parameter cut-off values determined by ROC curve analysis and (2) probabilistic logistic regression threshold-free analysis. Predicted infarct volumes and prediction errors based on four perfusion parameter maps were compared against observed infarcts. In 93 patients with successful recanalization, the mean observed infarct volume was 35.7 ± 61.9 ml (the reference for core infarct not savable by reperfusion). Optimal parameter thresholds predicted mean infarct volumes between 53.2 ± 44.4 and 125.0 ± 95.4 ml whereas threshold-free analysis predicted mean volumes between 35.9 ± 28.5 and 36.1 ± 29.0 ml. In 68 patients with persistent occlusion, the mean observed infarct volume was 113.4 ± 138.3 ml (the reference to define penumbral infarct savable by reperfusion). Predicted mean infarct volumes by parameter thresholds ranged from 91.4 ± 81.5 to 163.8 ± 135.7 ml, by threshold-free analysis from 113.2 ± 89.9 to 113.5 ± 89.0 ml. Threshold-free prediction of infarct volumes had a higher precision and lower patient-specific prediction error than conventional thresholding. Penumbra to core lesion mismatch estimate may therefore benefit from threshold-free CTP analysis.

Correction: Cerebral Hemodynamics in Patients with Hemolytic Uremic Syndrome Assessed by Susceptibility Weighted Imaging and Four-Dimensional Non-Contrast MR Angiography.

[This corrects the article DOI: 10.1371/journal.pone.0164863.].

Hypointense Vessels Detected by Susceptibility-Weighted Imaging Identifies Tissue at Risk of Infarction in Anterior Circulation Stroke.

BACKGROUND AND PURPOSE: The diagnostic value of susceptibility-weighted magnetic resonance imaging of acute stroke patients has shown potential as a surrogate marker of impaired hemodynamics. We investigate the value of asymmetrical hypointense cerebral vessels (HV) for the identification of vessel status and tissue at risk of infarction (TaR). METHODS: Symmetry of HV was visually rated on SWI data from a well-defined population of acute anterior circulation stroke with onset <24 hours. MRI perfusion data was analyzed and volumes of tissue at risk segmented using a delay threshold of Tmax> 6 seconds. Status of the extra- and intracranial arteries was assessed by ultrasound and MR angiography. RESULTS: 35 patients were included (12 women; median age 69 years, IQR 61-77; median NIHSS at admission 10, IQR 6-20). Asymmetrically distributed HV were detected at the stroke hemisphere in 25 patients (71%). Of those, 12 patients displayed occlusion of the middle cerebral artery, whereas occlusion of the extracranial ICA was detected in 6 patients. TaR was larger, yet not significantly different in patients with asymmetrically HV (mean volume 38.9 ml, SD 52.9 ml) compared to patients showing symmetrical HV (4.2 ml; SD 10.7 ml, p-value 0.081). Significant differences where, however, found after excluding patients with extracranial ICA occlusions (42.9 ml; SD 50.4 ml vs. 4.2 ml, SD 10.8 ml, p-value 0.025). CONCLUSION: Visual analysis of HV in SWI identifies tissue at risk in patients with anterior circulation stroke. Potentially pre-existing extracranial ICA occlusions leading to prominent HV have to be considered as a confounding factor.

Anisotropic cerebral vascular architecture causes orientation dependency in cerebral blood flow and volume measured with dynamic susceptibility contrast magnetic resonance imaging.

Measurements of cerebral perfusion using dynamic susceptibility contrast magnetic resonance imaging rely on the assumption of isotropic vascular architecture. However, a considerable fraction of vessels runs in parallel with white matter tracts. Here, we investigate the effects of tissue orientation on dynamic susceptibility contrast magnetic resonance imaging. Tissue orientation was measured using diffusion tensor imaging and dynamic susceptibility contrast was performed with gradient echo planar imaging. Perfusion parameters and the raw dynamic susceptibility contrast signals were correlated with tissue orientation. Additionally, numerical simulations were performed for a range of vascular volumes of both the isotropic vascular bed and anisotropic vessel components, as well as for a range of contrast agent concentrations. The effect of the contrast agent was much larger in white matter tissue perpendicular to the main magnetic field compared to white matter parallel to the main magnetic field. In addition, cerebral blood flow and cerebral blood volume were affected in the same way with angle-dependent variations of up to 130%. Mean transit time and time to maximum of the residual curve exhibited weak orientation dependency of 10%. Numerical simulations agreed with the measured data, showing that one-third of the white matter vascular volume is comprised of vessels running in parallel with the fibre tracts.

Cerebral Hemodynamics in Patients with Hemolytic Uremic Syndrome Assessed by Susceptibility Weighted Imaging and Four-Dimensional Non-Contrast MR Angiography.

BACKGROUND AND PURPOSE: Conventional magnetic resonance imaging (MRI) of patients with hemolytic uremic syndrome (HUS) and neurological symptoms performed during an epidemic outbreak of Escherichia coli O104:H4 in Northern Europe has previously shown pathological changes in only approximately 50% of patients. In contrast, susceptibility-weighted imaging (SWI) revealed a loss of venous contrast in a large number of patients. We hypothesized that this observation may be due to an increase in cerebral blood flow (CBF) and aimed to identify a plausible cause. MATERIALS AND METHODS: Baseline 1.5T MRI scans of 36 patients (female, 26; male, 10; mean age, 38.2±19.3 years) were evaluated. Venous contrast was rated on standard SWI minimum intensity projections. A prototype four-dimensional (time resolved) magnetic resonance angiography (4D MRA) assessed cerebral hemodynamics by global time-to-peak (TTP), as a surrogate marker for CBF. Clinical parameters studied were hemoglobin, hematocrit, creatinine, urea levels, blood pressure, heart rate, and end-tidal CO2. RESULTS: SWI venous contrast was abnormally low in 33 of 36 patients. TTP ranged from 3.7 to 10.2 frames (mean, 7.9 ± 1.4). Hemoglobin at the time of MRI (n = 35) was decreased in all patients (range, 5.0 to 12.6 g/dL; mean, 8.2 ± 1.4); hematocrit (n = 33) was abnormally low in all but a single patient (range, 14.3 to 37.2%; mean, 23.7 ± 4.2). Creatinine was abnormally high in 30 of 36 patients (83%) (range, 0.8 to 9.7; mean, 3.7 ± 2.2). SWI venous contrast correlated significantly with hemoglobin (r = 0.52, P = 0.0015), hematocrit (r = 0.65, P < 0.001), and TTP (r = 0.35, P = 0.036). No correlation of SWI with blood pressure, heart rate, end-tidal CO2, creatinine, and urea level was observed. Findings suggest that the loss of venous contrast is related to an increase in CBF secondary to severe anemia related to HUS. SWI contrast of patients with pathological conventional MRI findings was significantly lower compared to patients with normal MRI (mean SWI score, 1.41 and 2.05, respectively; P = 0.04). In patients with abnormal conventional MRI, mean TTP (7.45), mean hemoglobin (7.65), and mean hematocrit (22.0) were lower compared to patients with normal conventional MRI scans (mean TTP = 8.28, mean hemoglobin = 8.63, mean hematocrit = 25.23). CONCLUSION: In contrast to conventional MRI, almost all patients showed pathological changes in cerebral hemodynamics assessed by SWI and 4D MRA. Loss of venous contrast on SWI is most likely the result of an increase in CBF and may be related to the acute onset of anemia. Future studies will be needed to assess a possible therapeutic effect of blood transfusions in patients with HUS and neurological symptoms.

Classifiers for Ischemic Stroke Lesion Segmentation: A Comparison Study.

MOTIVATION: Ischemic stroke, triggered by an obstruction in the cerebral blood supply, leads to infarction of the affected brain tissue. An accurate and reproducible automatic segmentation is of high interest, since the lesion volume is an important end-point for clinical trials. However, various factors, such as the high variance in lesion shape, location and appearance, render it a difficult task. METHODS: In this article, nine classification methods (e.g. Generalized Linear Models, Random Decision Forests and Convolutional Neural Networks) are evaluated and compared with each other using 37 multiparametric MRI datasets of ischemic stroke patients in the sub-acute phase in terms of their accuracy and reliability for ischemic stroke lesion segmentation. Within this context, a multi-spectral classification approach is compared against mono-spectral classification performance using only FLAIR MRI datasets and two sets of expert segmentations are used for inter-observer agreement evaluation. RESULTS AND CONCLUSION: The results of this study reveal that high-level machine learning methods lead to significantly better segmentation results compared to the rather simple classification methods, pointing towards a difficult non-linear problem. The overall best segmentation results were achieved by a Random Decision Forest and a Convolutional Neural Networks classification approach, even outperforming all previously published results. However, none of the methods tested in this work are capable of achieving results in the range of the human observer agreement and the automatic ischemic stroke lesion segmentation remains a complicated problem that needs to be explored in more detail to improve the segmentation results.