Imaging biomarkers of cerebral edema automatically extracted from routine CT scans of large vessel occlusion strokes.

BACKGROUND AND PURPOSE: Volumetric and densitometric biomarkers have been proposed to better quantify cerebral edema after stroke, but their relative performance has not been rigorously evaluated. METHODS: Patients with large vessel occlusion stroke from three institutions were analyzed. An automated pipeline extracted brain, cerebrospinal fluid (CSF), and infarct volumes from serial CTs. Several biomarkers were measured: change in global CSF volume from baseline (ΔCSF); ratio of CSF volumes between hemispheres (CSF ratio); and relative density of infarct region compared with mirrored contralateral region (net water uptake [NWU]). These were compared to radiographic standards, midline shift and relative hemispheric volume (RHV) and malignant edema, defined as deterioration resulting in need for osmotic therapy, decompressive surgery, or death. RESULTS: We analyzed 255 patients with 210 baseline CTs, 255 24-hour CTs, and 81 72-hour CTs. Of these, 35 (14%) developed malignant edema and 63 (27%) midline shift. CSF metrics could be calculated for 310 (92%), while NWU could only be obtained from 193 (57%). Peak midline shift was correlated with baseline CSF ratio (ρ = -.22) and with CSF ratio and ΔCSF at 24 hours (ρ = -.55/.63) and 72 hours (ρ = -.66/.69), but not with NWU (ρ = .15/.25). Similarly, CSF ratio was correlated with RHV (ρ = -.69/-.78), while NWU was not. Adjusting for age, National Institutes of Health Stroke Scale, tissue plasminogen activator treatment, and Alberta Stroke Program Early CT Score, CSF ratio (odds ratio [OR]: 1.95 per 0.1, 95% confidence interval [CI]: 1.52-2.59) and ΔCSF at 24 hours (OR: 1.87 per 10%, 95% CI: 1.47-2.49) were associated with malignant edema. CONCLUSION: CSF volumetric biomarkers can be automatically measured from almost all routine CTs and correlate better with standard edema endpoints than net water uptake.

Normalization of cerebral hemodynamics after hematopoietic stem cell transplant in children with sickle cell disease.

Children with sickle cell disease (SCD) demonstrate cerebral hemodynamic stress and are at high risk of strokes. We hypothesized that curative hematopoietic stem cell transplant (HSCT) normalizes cerebral hemodynamics in children with SCD compared with pre-transplant baseline. Whole-brain cerebral blood flow (CBF) and oxygen extraction fraction (OEF) were measured by magnetic resonance imaging 1 to 3 months before and 12 to 24 months after HSCT in 10 children with SCD. Three children had prior overt strokes, 5 children had prior silent strokes, and 1 child had abnormal transcranial Doppler ultrasound velocities. CBF and OEF of HSCT recipients were compared with non-SCD control participants and with SCD participants receiving chronic red blood cell transfusion therapy (CRTT) before and after a scheduled transfusion. Seven participants received matched sibling donor HSCT, and 3 participants received 8 out of 8 matched unrelated donor HSCT. All received reduced-intensity preparation and maintained engraftment, free of hemolytic anemia and SCD symptoms. Pre-transplant, CBF (93.5 mL/100 g/min) and OEF (36.8%) were elevated compared with non-SCD control participants, declining significantly 1 to 2 years after HSCT (CBF, 72.7 mL/100 g per minute; P = .004; OEF, 27.0%; P = .002), with post-HSCT CBF and OEF similar to non-SCD control participants. Furthermore, HSCT recipients demonstrated greater reduction in CBF (-19.4 mL/100 g/min) and OEF (-8.1%) after HSCT than children with SCD receiving CRTT after a scheduled transfusion (CBF, -0.9 mL/100 g/min; P = .024; OEF, -3.3%; P = .001). Curative HSCT normalizes whole-brain hemodynamics in children with SCD. This restoration of cerebral oxygen reserve may explain stroke protection after HSCT in this high-risk patient population.

Accelerating Prediction of Malignant Cerebral Edema After Ischemic Stroke with Automated Image Analysis and Explainable Neural Networks.

BACKGROUND: Malignant cerebral edema is a devastating complication of stroke, resulting in deterioration and death if hemicraniectomy is not performed prior to herniation. Current approaches for predicting this relatively rare complication often require advanced imaging and still suffer from suboptimal performance. We performed a pilot study to evaluate whether neural networks incorporating data extracted from routine computed tomography (CT) imaging could enhance prediction of edema in a large diverse stroke cohort. METHODS: An automated imaging pipeline retrospectively extracted volumetric data, including cerebrospinal fluid (CSF) volumes and the hemispheric CSF volume ratio, from baseline and 24 h CT scans performed in participants of an international stroke cohort study. Fully connected and long short-term memory (LSTM) neural networks were trained using serial clinical and imaging data to predict those who would require hemicraniectomy or die with midline shift. The performance of these models was tested, in comparison with regression models and the Enhanced Detection of Edema in Malignant Anterior Circulation Stroke (EDEMA) score, using cross-validation to construct precision-recall curves. RESULTS: Twenty of 598 patients developed malignant edema (12 required surgery, 8 died). The regression model provided 95% recall but only 32% precision (area under the precision-recall curve [AUPRC] 0.74), similar to the EDEMA score (precision 28%, AUPRC 0.66). The fully connected network did not perform better (precision 33%, AUPRC 0.71), but the LSTM model provided 100% recall and 87% precision (AUPRC 0.97) in the overall cohort and the subgroup with a National Institutes of Health Stroke Scale (NIHSS) score ≥ 8 (p = 0.0001 vs. regression and fully connected models). Features providing the most predictive importance were the hemispheric CSF ratio and NIHSS score measured at 24 h. CONCLUSIONS: An LSTM neural network incorporating volumetric data extracted from routine CT scans identified all cases of malignant cerebral edema by 24 h after stroke, with significantly fewer false positives than a fully connected neural network, regression model, and the validated EDEMA score. This preliminary work requires prospective validation but provides proof of principle that a deep learning framework could assist in selecting patients for surgery prior to deterioration.

Hemispheric CSF volume ratio quantifies progression and severity of cerebral edema after acute hemispheric stroke.

As swelling occurs, CSF is preferentially displaced from the ischemic hemisphere. The ratio of CSF volume in the stroke-affected hemisphere to that in the contralateral hemisphere may quantify the progression of cerebral edema. We automatically segmented CSF from 1,875 routine CTs performed within 96 hours of stroke onset in 924 participants of a stroke cohort study. In 737 subjects with follow-up imaging beyond 24-hours, edema severity was classified as affecting less than one-third of the hemisphere (CED-1), large hemispheric infarction (LHI, over one-third the hemisphere), without midline shift (CED-2) or with midline shift (CED-3). Malignant edema was LHI resulting in deterioration, requiring osmotic therapy, surgery, or resulting in death. Hemispheric CSF ratio was lower on baseline CT in those with LHI (0.91 vs. 0.97, p < 0.0001) and decreased more rapidly in those with LHI who developed midline shift (0.01 per hour for CED-3 vs. 0.004/hour CED-2). The ratio at 24-hours was lower in those with midline shift (0.41, IQR 0.30-0.57 vs. 0.66, 0.56-0.81 for CED-2). A ratio below 0.50 provided 90% sensitivity, 82% specificity for predicting malignant edema among those with LHI (AUC 0.91, 0.85-0.96). This suggests that the hemispheric CSF ratio may provide an accessible early biomarker of edema severity.

Quantitative Serial CT Imaging-Derived Features Improve Prediction of Malignant Cerebral Edema after Ischemic Stroke.

INTRODUCTION: Malignant cerebral edema develops in a small subset of patients with hemispheric strokes, precipitating deterioration and death if decompressive hemicraniectomy (DHC) is not performed in a timely manner. Predicting which stroke patients will develop malignant edema is imprecise based on clinical data alone. Head computed tomography (CT) imaging is often performed at baseline and 24-h. We determined the incremental value of incorporating imaging-derived features from serial CTs to enhance prediction of malignant edema. METHODS: We identified hemispheric stroke patients at three sites with NIHSS ≥ 7 who had baseline as well as 24-h clinical and CT imaging data. We extracted quantitative imaging features from baseline and follow-up CTs, including CSF volume, intracranial reserve (CSF/cranial volume), as well as midline shift (MLS) and infarct-related hypodensity volume. Potentially lethal malignant edema was defined as requiring DHC or dying with MLS over 5-mm. We built machine-learning models using logistic regression first with baseline data and then adding 24-h data including reduction in CSF volume (ΔCSF). Model performance was evaluated with cross-validation using metrics of recall (sensitivity), precision (predictive value), as well as area under receiver-operating-characteristic and precision-recall curves (AUROC, AUPRC). RESULTS: Twenty of 361 patients (6%) died or underwent DHC. Baseline clinical variables alone had recall of 60% with low precision (7%), AUROC 0.59, AUPRC 0.15. Adding baseline intracranial reserve improved recall to 80% and AUROC to 0.82 but precision remained only 16% (AUPRC 0.28). Incorporating ΔCSF improved AUPRC to 0.53 (AUROC 0.91) while all imaging features further improved prediction (recall 90%, precision 38%, AUROC 0.96, AUPRC 0.66). CONCLUSION: Incorporating quantitative CT-based imaging features from baseline and 24-h CT enhances identification of patients with malignant edema needing DHC. Further refinements and external validation of such imaging-based machine-learning models are required.

Application of Machine Learning to Automated Analysis of Cerebral Edema in Large Cohorts of Ischemic Stroke Patients.

Cerebral edema contributes to neurological deterioration and death after hemispheric stroke but there remains no effective means of preventing or accurately predicting its occurrence. Big data approaches may provide insights into the biologic variability and genetic contributions to severity and time course of cerebral edema. These methods require quantitative analyses of edema severity across large cohorts of stroke patients. We have proposed that changes in cerebrospinal fluid (CSF) volume over time may represent a sensitive and dynamic marker of edema progression that can be measured from routinely available CT scans. To facilitate and scale up such approaches we have created a machine learning algorithm capable of segmenting and measuring CSF volume from serial CT scans of stroke patients. We now present results of our preliminary processing pipeline that was able to efficiently extract CSF volumetrics from an initial cohort of 155 subjects enrolled in a prospective longitudinal stroke study. We demonstrate a high degree of reproducibility in total cranial volume registration between scans (R = 0.982) as well as a strong correlation of baseline CSF volume and patient age (as a surrogate of brain atrophy, R = 0.725). Reduction in CSF volume from baseline to final CT was correlated with infarct volume (R = 0.715) and degree of midline shift (quadratic model, p < 2.2 × 10-16). We utilized generalized estimating equations (GEE) to model CSF volumes over time (using linear and quadratic terms), adjusting for age. This model demonstrated that CSF volume decreases over time (p < 2.2 × 10-13) and is lower in those with cerebral edema (p = 0.0004). We are now fully automating this pipeline to allow rapid analysis of even larger cohorts of stroke patients from multiple sites using an XNAT (eXtensible Neuroimaging Archive Toolkit) platform. Data on kinetics of edema across thousands of patients will facilitate precision approaches to prediction of malignant edema as well as modeling of variability and further understanding of genetic variants that influence edema severity.

CAPTURE: Consistently Acquired Projections for Tuned and Robust Estimation: A Self-Navigated Respiratory Motion Correction Approach.

OBJECTIVES: In this study, we present a fully automated and robust self-navigated approach to obtain 4-dimensional (4-D) motion-resolved images during free breathing. MATERIALS AND METHODS: The proposed method, Consistently Acquired Projections for Tuned and Robust Estimation (CAPTURE), is a variant of the stack-of-stars gradient-echo sequence. A 1-D navigator was consistently acquired at a fixed azimuthal angle for all stacks of spokes to reduce nonphysiological signal contamination due to system imperfections. The resulting projections were then "tuned" using complex phase rotation to adapt to scan-to-scan variations, followed by the detection of the respiratory curve. Four-dimensional motion-corrected and uncorrected images were then reconstructed via respiratory and temporal binning, respectively.This Health Insurance Portability and Accountability Act-compliant study was performed with Institutional Review Board approval. A phantom experiment was performed using a custom-made deformable motion phantom with an adjustable frequency and amplitude. For in vivo experiments, 10 healthy participants and 12 liver tumor patients provided informed consent and were imaged with the CAPTURE sequence.Two radiologists, blinded to which images were motion-corrected and which were not, independently reviewed the images and scored the image quality using a 5-point Likert scale. RESULTS: In the respiratory motion phantom experiment, CAPTURE reversed the effects of motion blurring and restored edge sharpness from 36% to 78% of that observed in the images from the static scan.Despite large intra- and intersubject variability in respiration patterns, CAPTURE successfully detected the respiratory motion signal in all participants and significantly improved the image quality according to the subjective radiological assessments of 2 raters (P < 0.05 for both raters) with a 1 to 2-point improvement in the median Likert scores across the whole set of participants. Small lesions (<1 cm in size) which might otherwise be missed on uncorrected images because of motion blurring were more clearly depicted on the CAPTURE images. CONCLUSIONS: CAPTURE provides a robust and fully automated solution for obtaining 4-D motion-resolved images in a free-breathing setting. With its unique tuning feature, CAPTURE can adapt to large intersubject and interscan variations. CAPTURE also enables better lesion delineation because of improved image sharpness, thereby increasing the visibility of small lesions.

Large-Vessel Vasculopathy in Children With Sickle Cell Disease: A Magnetic Resonance Imaging Study of Infarct Topography and Focal Atrophy.

BACKGROUND: Large-vessel vasculopathy (LVV) increases stroke risk in pediatric sickle cell disease beyond the baseline elevated stroke risk in this vulnerable population. The mechanisms underlying this added risk and its unique impact on the developing brain are not established. METHODS: We analyzed magnetic resonance imaging and angiography scans of 66 children with sickle cell disease and infarcts by infarct density heatmaps and Jacobian determinants, a metric utilized to delineate focal volume change, to investigate if infarct location, volume, frequency, and cerebral atrophy differed among hemispheres with and without LVV. RESULTS: Infarct density heatmaps demonstrated infarct "hot spots" within the deep white matter internal border zone region in both LVV and non-LVV hemispheres, but with greater infarct density and larger infarct volumes in LVV hemispheres (2.2 mL versus 0.25 mL, P < 0.001). Additional scattered cortical infarcts in the internal carotid artery territory occurred in LVV hemispheres, but were rare in non-LVV hemispheres. Jacobian determinants revealed greater atrophy in gray and white matter of the parietal lobes of LVV compared with non-LVV hemispheres. CONCLUSION: Large-vessel vasculopathy in sickle cell disease appears to increase ischemic vulnerability in the borderzone region, as demonstrated by the increased frequency and extent of infarction within deep white matter, and increased risk of focal atrophy. Scattered infarctions across the LVV-affected hemispheres suggest additional stroke etiologies of vasculopathy (i.e., thromboembolism) in addition to chronic hypoxia-ischemia.

CSF Volumetric Analysis for Quantification of Cerebral Edema After Hemispheric Infarction.

BACKGROUND: Malignant cerebral edema (CED) complicates at least 20 % of large hemispheric infarcts (LHI) and may result in neurological deterioration or death. Midline shift (MLS) is a standard but crude measure of edema severity. We propose that volumetric analysis of shifts in cerebrospinal fluid (CSF) over time provides a reliable means of quantifying the spectrum of edema severity after LHI. METHODS: We identified 38 patients from 2008 to 2014 with NIHSS ≥8, baseline CT <6 h after stroke onset, at least 1 follow-up (FU) CT, and no parenchymal hematoma. The volumes of CSF (sulci, ventricles, and cisterns) ipsilateral (IL) and contralateral (CL) to infarct on baseline and FU CTs were quantified by manually assisted outlining with MIPAV image analysis software, as was infarct volume and MLS on FU CTs. Percentage change in CSF volumes (∆CSF) from baseline to FU scans was correlated with MLS and compared in those with vs. without malignant edema (defined as hemicraniectomy, osmotic therapy, or death/neurological deterioration with MLS ≥5 mm). RESULTS: 11 of 38 subjects (29 %) developed malignant edema. Neither baseline NIHSS nor CSF volume differed between those with and without edema (median NIHSS 18 vs. 13, p = 0.12, CSF volume 102 vs. 124 ml, p = 0.16). Inter-rater reliability for CSF measurements was excellent (intraclass correlation coefficient 0.97). ∆CSF correlated strongly with MLS at peak edema (r = -0.75), even adjusting for infarct volume (p = 0.009). ∆CSF was also greater in those with malignant edema [-55 % (IQR -49 to -62) vs. -36 % (-27 to -45), p = 0.004]. ∆CSF was the greatest within IL sulci [-97 % (-86 to -99) vs. -71 % (-41 to -79), p = 0.002] but also significantly greater within CL sulci in those with malignant edema [-50 % (-29 to -65) vs. -25 % (0 to -31), p = 0.014]. More than half this CSF volume reduction occurred by the time of first FU CT around 24 h after stroke, while MLS rose later. CONCLUSIONS: Volumetric CSF analysis reliably quantifies CED and distinguishes those with malignant edema and MLS from those with a more benign course after LHI. ∆CSF may provide an earlier and more sensitive indicator of edema severity across a broader dynamic range than MLS.

Simulation of brain mass effect with an arbitrary Lagrangian and Eulerian FEM.

Estimation of intracranial stress distribution caused by mass effect is critical to the management of hemorrhagic stroke or brain tumor patients, who may suffer severe secondary brain injury from brain tissue compression. Coupling with physiological parameters that are readily available using MRI, eg, tissue perfusion, a non-invasive, quantitative and regional estimation of intracranial stress distribution could offer a better understanding of brain tissue's reaction under mass effect. A quantitative and sound measurement serving this particular purpose remains elusive due to multiple challenges associated with biomechanical modeling of the brain. One such challenge for the conventional Lagrangian frame based finite element method (LFEM) is that the mesh distortion resulted from the expansion of the mass effects can terminate the simulation prematurely before the desired pressure loading is achieved. In this work, we adopted an arbitrary Lagrangian and Eulerian FEM method (ALEF) with explicit dynamic solutions to simulate the expansion of brain mass effects caused by a pressure loading. This approach consists of three phases: 1) a Lagrangian phase to deform mesh like LFEM, 2) a mesh smoothing phase to reduce mesh distortion, and 3) an Eulerian phase to map the state variables from the old mesh to the smoothed one. In 2D simulations with simulated geometries, this approach is able to model substantially larger deformations compared to LFEM. We further applied this approach to a simulation with 3D real brain geometry to quantify the distribution of von Mises stress within the brain.