Automated multiclass segmentation, quantification, and visualization of the diseased aorta on hybrid PET/CT-SEQUOIA.

BACKGROUND: Cardiovascular disease is the most common cause of death worldwide, including infection and inflammation related conditions. Multiple studies have demonstrated potential advantages of hybrid positron emission tomography combined with computed tomography (PET/CT) as an adjunct to current clinical inflammatory and infectious biochemical markers. To quantitatively analyze vascular diseases at PET/CT, robust segmentation of the aorta is necessary. However, manual segmentation is extremely time-consuming and labor-intensive. PURPOSE: To investigate the feasibility and accuracy of an automated tool to segment and quantify multiple parts of the diseased aorta on unenhanced low-dose computed tomography (LDCT) as an anatomical reference for PET-assessed vascular disease. METHODS: A software pipeline was developed including automated segmentation using a 3D U-Net, calcium scoring, PET uptake quantification, background measurement, radiomics feature extraction, and 2D surface visualization of vessel wall calcium and tracer uptake distribution. To train the 3D U-Net, 352 non-contrast LDCTs from (2-[18 F]FDG and Na[18 F]F) PET/CTs performed in patients with various vascular pathologies with manual segmentation of the ascending aorta, aortic arch, descending aorta, and abdominal aorta were used. The last 22 consecutive scans were used as a hold-out internal test set. The remaining dataset was randomly split into training (n = 264; 80%) and validation (n = 66; 20%) sets. Further evaluation was performed on an external test set of 49 PET/CTs. The dice similarity coefficient (DSC) and Hausdorff distance (HD) were used to assess segmentation performance. Automatically obtained calcium scores and uptake values were compared with manual scoring obtained using clinical softwares (syngo.via and Affinity Viewer) in six patient images. intraclass correlation coefficients (ICC) were calculated to validate calcium and uptake values. RESULTS: Fully automated segmentation of the aorta using a 3D U-Net was feasible in LDCT obtained from PET/CT scans. The external test set yielded a DSC of 0.867 ± 0.030 and HD of 1.0 [0.6-1.4] mm, similar to an open-source model with a DSC of 0.864 ± 0.023 and HD of 1.4 [1.0-1.8] mm. Quantification of calcium and uptake values were in excellent agreement with clinical software (ICC: 1.00 [1.00-1.00] and 0.99 [0.93-1.00] for calcium and uptake values, respectively). CONCLUSIONS: We present an automated pipeline to segment the ascending aorta, aortic arch, descending aorta, and abdominal aorta on LDCT from PET/CT and to accurately provide uptake values, calcium scores, background measurement, radiomics features, and a 2D visualization. We call this algorithm SEQUOIA (SEgmentation, QUantification, and visualizatiOn of the dIseased Aorta) and is available at https://github.com/UMCG-CVI/SEQUOIA. This model could augment the utility of aortic evaluation at PET/CT studies tremendously, irrespective of the tracer, and potentially provide fast and reliable quantification of cardiovascular diseases in clinical practice, both for primary diagnosis and disease monitoring.

Prediction of pseudoprogression in post-treatment glioblastoma using dynamic susceptibility contrast-derived oxygenation and microvascular transit time heterogeneity measures.

OBJECTIVES: To evaluate the added value of MR dynamic susceptibility contrast (DSC)-perfusion-weighted imaging (PWI)-derived tumour microvascular and oxygenation information with cerebral blood volume (CBV) to distinguish pseudoprogression from true progression (TP) in post-treatment glioblastoma. METHODS: This retrospective single-institution study included patients with isocitrate dehydrogenase (IDH) wild-type glioblastoma and a newly developed or enlarging measurable contrast-enhancing mass within 12 weeks after concurrent chemoradiotherapy. CBV, capillary transit time heterogeneity (CTH), oxygen extraction fraction (OEF), and cerebral metabolic rate of oxygen (CMRO2) were obtained from DSC-PWI. Predictors were selected using univariable logistic regression, and performance was measured with adjusted diagnostic odds with tumour volume and area under the curve (AUC) of receiver operating characteristics analysis. RESULTS: A total of 103 patients were included (mean age, 59.6 years; 59 women), with 67 cases of TP and 36 cases of pseudoprogression. Pseudoprogression exhibited higher CTH (4.0 vs. 3.4, p = .019) and higher OEF (12.7 vs. 10.7, p = .014) than TP, but a similar CBV (1.48 vs. 1.53, p = .13) and CMRO2 (7.7 vs. 7.3s, p = .598). Independent of tumour volume, both high CTH (adjusted odds ratio [OR] 1.52; 95% confidence interval [CI]: 1.11-2.09, p = .009) and high OEF (adjusted OR 1.17; 95% CI:1.03-1.33, p = .016) were predictors of pseudoprogression. The combination of CTH, OEF, and CBV yielded higher diagnostic performance (AUC 0.71) than CBV alone (AUC 0.65). CONCLUSION: High intratumoural capillary transit heterogeneity and high oxygen extraction fraction derived from DSC-PWI have enhanced the diagnostic value of CBV in pseudoprogression of post-treatment IDH-wild type glioblastoma. CLINICAL RELEVANCE STATEMENT: In the early post-treatment stage of glioblastoma, pseudoprogression exhibited both high oxygen extraction fraction and high capillary transit heterogeneity and these dynamic susceptibility contrast-perfusion weighted imaging derived parameters have added value in cerebral blood volume-based noninvasive differentiation of pseudoprogression from true progression. KEY POINTS: • Capillary transit time heterogeneity and oxygen extraction fraction can be measured noninvasively through processing of dynamic susceptibility contrast imaging. • Pseudoprogression exhibited higher capillary transit time heterogeneity and higher oxygen extraction fraction than true progression. • A combination of cerebral blood volume, capillary transit time heterogeneity, and oxygen extraction fraction yielded the highest diagnostic performance (area under the curve 0.71).

Hematopoietic stem-cell gene therapy is associated with restored white matter microvascular function in cerebral adrenoleukodystrophy.

Blood-brain barrier disruption marks the onset of cerebral adrenoleukodystrophy (CALD), a devastating cerebral demyelinating disease caused by loss of ABCD1 gene function. The underlying mechanism are not well understood, but evidence suggests that microvascular dysfunction is involved. We analyzed cerebral perfusion imaging in boys with CALD treated with autologous hematopoietic stem-cells transduced with the Lenti-D lentiviral vector that contains ABCD1 cDNA as part of a single group, open-label phase 2-3 safety and efficacy study (NCT01896102) and patients treated with allogeneic hematopoietic stem cell transplantation. We found widespread and sustained normalization of white matter permeability and microvascular flow. We demonstrate that ABCD1 functional bone marrow-derived cells can engraft in the cerebral vascular and perivascular space. Inverse correlation between gene dosage and lesion growth suggests that corrected cells contribute long-term to remodeling of brain microvascular function. Further studies are needed to explore the longevity of these effects.

Combining Hepatic and Splenic CT Radiomic Features Improves Radiomic Analysis Performance for Liver Fibrosis Staging.

Background: The exact focus of computed tomography (CT)-based artificial intelligence techniques when staging liver fibrosis is still not exactly known. This study aimed to determine both the added value of splenic information to hepatic information, and the correlation between important radiomic features and information exploited by deep learning models for liver fibrosis staging by CT-based radiomics. Methods: The study design is retrospective. Radiomic features were extracted from both liver and spleen on portal venous phase CT images of 252 consecutive patients with histologically proven liver fibrosis stages between 2006 and 2018. The radiomics analyses for liver fibrosis staging were done by hepatic and hepatic-splenic features, respectively. The most predictive radiomic features were automatically selected by machine learning models. Results: When using splenic-hepatic features in the CT-based radiomics analysis, the average accuracy rates for significant fibrosis, advanced fibrosis, and cirrhosis were 88%, 82%, and 86%, and area under the receiver operating characteristic curves (AUCs) were 0.92, 0.81, and 0.85. The AUC of hepatic-splenic-based radiomics analysis with the ensemble classifier was 7% larger than that of hepatic-based analysis (p < 0.05). The most important features selected by machine learning models included both hepatic and splenic features, and they were consistent with the location maps indicating the focus of deep learning when predicting liver fibrosis stage. Conclusions: Adding CT-based splenic radiomic features to hepatic radiomic features increases radiomics analysis performance for liver fibrosis staging. The most important features of the radiomics analysis were consistent with the information exploited by deep learning.

Liver fibrosis staging by deep learning: a visual-based explanation of diagnostic decisions of the model.

OBJECTIVES: Deep learning has been proven to be able to stage liver fibrosis based on contrast-enhanced CT images. However, until now, the algorithm is used as a black box and lacks transparency. This study aimed to provide a visual-based explanation of the diagnostic decisions made by deep learning. METHODS: The liver fibrosis staging network (LFS network) was developed at contrast-enhanced CT images in the portal venous phase in 252 patients with histologically proven liver fibrosis stage. To give a visual explanation of the diagnostic decisions made by the LFS network, Gradient-weighted Class Activation Mapping (Grad-cam) was used to produce location maps indicating where the LFS network focuses on when predicting liver fibrosis stage. RESULTS: The LFS network had areas under the receiver operating characteristic curve of 0.92, 0.89, and 0.88 for staging significant fibrosis (F2-F4), advanced fibrosis (F3-F4), and cirrhosis (F4), respectively, on the test set. The location maps indicated that the LFS network had more focus on the liver surface in patients without liver fibrosis (F0), while it focused more on the parenchyma of the liver and spleen in case of cirrhosis (F4). CONCLUSIONS: Deep learning methods are able to exploit CT-based information from the liver surface, liver parenchyma, and extrahepatic information to predict liver fibrosis stage. Therefore, we suggest using the entire upper abdomen on CT images when developing deep learning-based liver fibrosis staging algorithms. KEY POINTS: • Deep learning algorithms can stage liver fibrosis using contrast-enhanced CT images, but the algorithm is still used as a black box and lacks transparency. • Location maps produced by Gradient-weighted Class Activation Mapping can indicate the focus of the liver fibrosis staging network. • Deep learning methods use CT-based information from the liver surface, liver parenchyma, and extrahepatic information to predict liver fibrosis stage.

Hemodynamic impairments within individual watershed areas in asymptomatic carotid artery stenosis by multimodal MRI.

Improved understanding of complex hemodynamic impairments in asymptomatic internal carotid artery stenosis (ICAS) is crucial to better assess stroke risks. Multimodal MRI is ideal for measuring brain hemodynamics and has the potential to improve diagnostics and treatment selections. We applied MRI-based perfusion and oxygenation-sensitive imaging in ICAS with the hypothesis that the sensitivity to hemodynamic impairments will improve within individual watershed areas (iWSA). We studied cerebral blood flow (CBF), cerebrovascular reactivity (CVR), relative cerebral blood volume (rCBV), relative oxygen extraction fraction (rOEF), oxygen extraction capacity (OEC) and capillary transit-time heterogeneity (CTH) in 29 patients with asymptomatic, unilateral ICAS (age 70.3 ± 7.0 y) and 30 age-matched healthy controls. In ICAS, we found significant impairments of CBF, CVR, rCBV, OEC, and CTH (strongest lateralization ΔCVR = -24%), but not of rOEF. Although the spatial overlap of compromised hemodynamic parameters within each patient varied in a complex manner, most pronounced changes of CBF, CVR and rCBV were detected within iWSAs (strongest effect ΔCVR = +117%). At the same time, CTH impairments were iWSA independent, indicating widespread dysfunction of capillary-level oxygen diffusivity. In summary, complementary MRI-based perfusion and oxygenation parameters offer deeper perspectives on complex microvascular impairments in individual patients. Furthermore, knowledge about iWSAs improves the sensitivity to hemodynamic impairments.

Theophylline as an Add-On to Thrombolytic Therapy in Acute Ischemic Stroke: A Randomized Placebo-Controlled Trial.

BACKGROUND AND PURPOSE: Delayed recanalization increases the risk of infarct growth and poor clinical outcome in acute ischemic stroke. The vasoactive agent theophylline has shown neuroprotective effects in animal stroke models but inconclusive results in case series and randomized clinical trials. The primary objective of this study was to evaluate whether theophylline, as an add-on to thrombolytic therapy, is safe and effective in acute ischemic stroke patients. METHODS: The TEA-Stroke trial (The Theophylline in Acute Ischemic Stroke) was an investigator-initiated 2-center, proof-of-concept, phase II clinical study with a randomized, double-blinded, placebo-controlled design. The main inclusion criteria were magnetic resonance imaging-verified acute ischemic stroke, moderate to severe neurological deficit (National Institutes of Health Stroke Scale score of ≥4), and treatment with thrombolysis within 4.5 hours of onset. Participants were randomly assigned in the ratio 1:1 to either 220 mg of intravenous theophylline or placebo. The co-primary outcomes were early clinical improvement on the National Institutes of Health Stroke Scale score and infarct growth on magnetic resonance imaging at 24-hour follow-up. RESULTS: Theophylline as an add-on to thrombolytic therapy improved the National Institutes of Health Stroke Scale score at 24 hours by mean 4.7 points (SD, 5.6) compared with an improvement of 1.3 points (SD, 7.5) in the control group (P=0.044). Mean infarct growth was 141.6% (SD, 126.5) and 104.1% (SD, 62.5) in the theophylline and control groups, respectively (P=0.146). Functional independence at 90 days was 61% in the theophylline group and 58% in the control group (P=0.802). CONCLUSIONS: This proof-of-concept trial investigated theophylline administration as an add-on to thrombolytic therapy in acute ischemic stroke. The co-primary end points early clinical improvement and infarct growth at 24-hour follow-up were not significantly different after post hoc correction for multiplicity (Bonferroni technique). The small study size precludes a conclusion as to whether theophylline has a neuroprotective effect but provides a promising clinical signal that may support a future clinical trial. Registration: URL: https://www.clinicaltrials.gov. Unique identifier: EudraCT number 2013-001989-42.

Attenuation of dopamine-induced GABA release in problem gamblers.

INTRODUCTION: We have previously shown that an interaction between medial prefrontal and parietal cortices is instrumental in promoting self-awareness via synchronizing oscillations in the gamma range. The synchronization of these oscillations is modulated by dopamine release. Given that such oscillations result from intermittent GABA stimulation of pyramidal cells, it is of interest to determine whether the dopaminergic system regulates GABA release directly in cortical paralimbic regions. Here, we test the hypothesis that the regulation of the GABA-ergic system by the dopaminergic system becomes attenuated in problem gamblers resulting in addictive behaviors and impaired self-awareness. METHODS: [11 C]Ro15-4513 PET, a marker of benzodiazepine α1/α5 receptor availability in the GABA receptor complex, was used to detect changes in synaptic GABA levels after oral doses of 100mg L-dopa in a double-blind controlled study of male problem gamblers (N = 10) and age-matched healthy male controls (N = 10). RESULTS: The mean reduction of cortical gray matter GABA/BDZ receptor availability induced by L-dopa was significantly attenuated in the problem gambling group compared to the healthy control group (p = 0.0377). CONCLUSIONS: Our findings demonstrate that: (a) Exogenous dopamine can induce synaptic GABA release in healthy controls. (b) This release is attenuated in frontal cortical areas of males suffering from problem gambling, possibly contributing to their loss of inhibitory control. This suggests that dysfunctional dopamine regulation of GABA release may contribute to problem gambling and gambling disorder.

Robust estimation of hemo-dynamic parameters in traditional DCE-MRI models.

PURPOSE: In dynamic contrast enhanced (DCE) MRI, separation of signal contributions from perfusion and leakage requires robust estimation of parameters in a pharmacokinetic model. We present and quantify the performance of a method to compute tissue hemodynamic parameters from DCE data using established pharmacokinetic models. METHODS: We propose a Bayesian scheme to obtain perfusion metrics from DCE MRI data. Initial performance is assessed through digital phantoms of the extended Tofts model (ETM) and the two-compartment exchange model (2CXM), comparing the Bayesian scheme to the standard Levenberg-Marquardt (LM) algorithm. Digital phantoms are also invoked to identify limitations in the pharmacokinetic models related to measurement conditions. Using computed maps of the extra vascular volume (ve) from 19 glioma patients, we analyze differences in the number of un-physiological high-intensity ve values for both ETM and 2CXM, using a one-tailed paired t-test assuming un-equal variance. RESULTS: The Bayesian parameter estimation scheme demonstrated superior performance over the LM technique in the digital phantom simulations. In addition, we identified limitations in parameter reliability in relation to scan duration for the 2CXM. DCE data for glioma and cervical cancer patients was analyzed with both algorithms and demonstrated improvement in image readability for the Bayesian method. The Bayesian method demonstrated significantly fewer non-physiological high-intensity ve values for the ETM (p<0.0001) and the 2CXM (p<0.0001). CONCLUSION: We have demonstrated substantial improvement of the perceptive quality of pharmacokinetic parameters from advanced compartment models using the Bayesian parameter estimation scheme as compared to the LM technique.

Oxygenation differs among white matter hyperintensities, intersected fiber tracts and unaffected white matter.

White matter hyperintensities of presumed vascular origin are frequently observed on magnetic resonance imaging in normal aging. They are typically found in cerebral small vessel disease and suspected culprits in the etiology of complex age- and small vessel disease-related conditions, such as late-onset depression. White matter hyperintensities may interfere with surrounding white matter metabolic demands by disrupting fiber tract integrity. Meanwhile, risk factors for small vessel disease are thought to reduce tissue oxygenation, not only by reducing regional blood supply, but also by impairing capillary function. To address white matter oxygen supply-demand balance, we estimated voxel-wise capillary density as an index of resting white matter metabolism, and combined estimates of blood supply and capillary function to calculate white matter oxygen availability. We conducted a cross-sectional study with structural, perfusion- and diffusion-weighted magnetic resonance imaging in 21 patients with late-onset depression and 21 controls. We outlined white matter hyperintensities and used tractography to identify the tracts they intersect. Perfusion data comprised cerebral blood flow, blood volume, mean transit time and relative transit time heterogeneity-the latter a marker of capillary dysfunction. Based on these, white matter oxygenation was calculated as the steady state cerebral metabolic rate of oxygen under the assumption of normal tissue oxygen tension and vice versa. The number, volume and perfusion characteristics of white matter hyperintensities did not differ significantly between groups. Hemodynamic data showed white matter hyperintensities to have lower blood flow and blood volume, but higher relative transit time heterogeneity, than normal-appearing white matter, resulting in either reduced capillary metabolic rate of oxygen or oxygen tension. Intersected tracts showed significantly lower blood flow, blood volume and capillary metabolic rate of oxygen than normal-appearing white matter. Across groups, lower lesion oxygen tension was associated with higher lesion number and volume. Compared with normal-appearing white matter, tissue oxygenation is significantly reduced in white matter hyperintensities as well as the fiber tracts they intersect, independent of parallel late-onset depression. In white matter hyperintensities, reduced microvascular blood volume and concomitant capillary dysfunction indicate a severe oxygen supply-demand imbalance with hypoxic tissue injury. In intersected fiber tracts, parallel reductions in oxygenation and microvascular blood volume are consistent with adaptations to reduced metabolic demands. We speculate, that aging and vascular risk factors impair white matter hyperintensity perfusion and capillary function to create hypoxic tissue injury, which in turn affect the function and metabolic demands of the white matter tracts they disrupt.

Capillary flow disturbances after experimental subarachnoid hemorrhage: A contributor to delayed cerebral ischemia?

BACKGROUND: The high mortality and morbidity after SAH is partly due to DCI, which is traditionally ascribed to development of angiographic vasospasms. This relation has been challenged, and capillary flow disturbances are proposed as another mechanism contributing to brain damage after SAH. OBJECTIVE: To investigate capillary flow changes 4 days following experimental SAH. METHODS: SAH was induced by endovascular perforation of circle of Willis. We used TPM to evaluate blood flow characteristics. Cortical capillary diameters were investigated by both TPM and histology. RESULTS: We found elevated CTH and MTT of blood in SAH mice compared to sham animals. We observed capillaries with stagnant RBCs, and capillaries with increased RBC LD in the SAH group, suggesting severe blood maldistribution among cortical capillaries. Favoring that these capillary flow changes were primary to upstream vasoconstrictions, TPM showed no significant differences in arteriolar diameter between groups, while histological examination showed reduced capillary diameter in SAH group. CONCLUSION: Our study shows profound subacute hypoperfusion and capillary flow disturbances in a mouse SAH model and suggests that these changes are the result of changes in capillary function, rather than upstream vasospasm.

Bayesian modeling of Dynamic Contrast Enhanced MRI data in cerebral glioma patients improves the diagnostic quality of hemodynamic parameter maps.

PURPOSE: The purpose of this work is to investigate if the curve-fitting algorithm in Dynamic Contrast Enhanced (DCE) MRI experiments influences the diagnostic quality of calculated parameter maps. MATERIAL AND METHODS: We compared the Levenberg-Marquardt (LM) and a Bayesian method (BM) in DCE data of 42 glioma patients, using two compartmental models (extended Toft's and 2-compartment-exchange model). Logistic regression and an ordinal linear mixed model were used to investigate if the image quality differed between the curve-fitting algorithms and to quantify if image quality was affected for different parameters and algorithms. The diagnostic performance to discriminate between high-grade and low-grade gliomas was compared by applying a Wilcoxon signed-rank test (statistical significance p>0.05). Two neuroradiologists assessed different qualitative imaging features. RESULTS: Parameter maps based on BM, particularly those describing the blood-brain barrier, were superior those based on LM. The image quality was found to be significantly improved (p<0.001) for BM when assessed through independent clinical scores. In addition, given a set of clinical scores, the generating algorithm could be predicted with high accuracy (area under the receiver operating characteristic curve between 0.91 and 1). Using linear mixed models, image quality was found to be improved when applying the 2-compartment-exchange model compared to the extended Toft's model, regardless of the underlying fitting algorithm. Tumor grades were only differentiated reliably on plasma volume maps when applying BM. The curve-fitting algorithm had, however, no influence on grading when using parameter maps describing the blood-brain barrier. CONCLUSION: The Bayesian method has the potential to increase the diagnostic reliability of Dynamic Contrast Enhanced parameter maps in brain tumors. In our data, images based on the 2-compartment-exchange model were superior to those based on the extended Toft's model.

Better Diffusion Segmentation in Acute Ischemic Stroke Through Automatic Tree Learning Anomaly Segmentation.

Stroke is the second most common cause of death worldwide, responsible for 6.24 million deaths in 2015 (about 11% of all deaths). Three out of four stroke survivors suffer long term disability, as many cannot return to their prior employment or live independently. Eighty-seven percent of strokes are ischemic. As an increasing volume of ischemic brain tissue proceeds to permanent infarction in the hours following the onset, immediate treatment is pivotal to increase the likelihood of good clinical outcome for the patient. Triaging stroke patients for active therapy requires assessment of the volume of salvageable and irreversible damaged tissue, respectively. With Magnetic Resonance Imaging (MRI), diffusion-weighted imaging is commonly used to assess the extent of permanently damaged tissue, the core lesion. To speed up and standardize decision-making in acute stroke management we present a fully automated algorithm, ATLAS, for delineating the core lesion. We compare performance to widely used threshold based methodology, as well as a recently proposed state-of-the-art algorithm: COMBAT Stroke. ATLAS is a machine learning algorithm trained to match the lesion delineation by human experts. The algorithm utilizes decision trees along with spatial pre- and post-regularization to outline the lesion. As input data the algorithm takes images from 108 patients with acute anterior circulation stroke from the I-Know multicenter study. We divided the data into training and test data using leave-one-out cross validation to assess performance in independent patients. Performance was quantified by the Dice index. The median Dice coefficient of ATLAS algorithm was 0.6122, which was significantly higher than COMBAT Stroke, with a median Dice coefficient of 0.5636 (p < 0.0001) and the best possible performing methods based on thresholding of the diffusion weighted images (median Dice coefficient: 0.3951) or the apparent diffusion coefficient (median Dice coefficeint: 0.2839). Furthermore, the volume of the ATLAS segmentation was compared to the volume of the expert segmentation, yielding a standard deviation of the residuals of 10.25 ml compared to 17.53 ml for COMBAT Stroke. Since accurate quantification of the volume of permanently damaged tissue is essential in acute stroke patients, ATLAS may contribute to more optimal patient triaging for active or supportive therapy.

Prediction of Tissue Outcome and Assessment of Treatment Effect in Acute Ischemic Stroke Using Deep Learning.

BACKGROUND AND PURPOSE: Treatment options for patients with acute ischemic stroke depend on the volume of salvageable tissue. This volume assessment is currently based on fixed thresholds and single imagine modalities, limiting accuracy. We wish to develop and validate a predictive model capable of automatically identifying and combining acute imaging features to accurately predict final lesion volume. METHODS: Using acute magnetic resonance imaging, we developed and trained a deep convolutional neural network (CNNdeep) to predict final imaging outcome. A total of 222 patients were included, of which 187 were treated with rtPA (recombinant tissue-type plasminogen activator). The performance of CNNdeep was compared with a shallow CNN based on the perfusion-weighted imaging biomarker Tmax (CNNTmax), a shallow CNN based on a combination of 9 different biomarkers (CNNshallow), a generalized linear model, and thresholding of the diffusion-weighted imaging biomarker apparent diffusion coefficient (ADC) at 600×10-6 mm2/s (ADCthres). To assess whether CNNdeep is capable of differentiating outcomes of ±intravenous rtPA, patients not receiving intravenous rtPA were included to train CNNdeep,-rtpa to access a treatment effect. The networks' performances were evaluated using visual inspection, area under the receiver operating characteristic curve (AUC), and contrast. RESULTS: CNNdeep yields significantly better performance in predicting final outcome (AUC=0.88±0.12) than generalized linear model (AUC=0.78±0.12; P=0.005), CNNTmax (AUC=0.72±0.14; P<0.003), and ADCthres (AUC=0.66±0.13; P<0.0001) and a substantially better performance than CNNshallow (AUC=0.85±0.11; P=0.063). Measured by contrast, CNNdeep improves the predictions significantly, showing superiority to all other methods (P≤0.003). CNNdeep also seems to be able to differentiate outcomes based on treatment strategy with the volume of final infarct being significantly different (P=0.048). CONCLUSIONS: The considerable prediction improvement accuracy over current state of the art increases the potential for automated decision support in providing recommendations for personalized treatment plans.

Boosted Tree Model Reforms Multimodal Magnetic Resonance Imaging Infarct Prediction in Acute Stroke.

BACKGROUND AND PURPOSE: Stroke imaging is pivotal for diagnosis and stratification of patients with acute ischemic stroke to treatment. The potential of combining multimodal information into reliable estimates of outcome learning calls for robust machine learning techniques with high flexibility and accuracy. We applied the novel extreme gradient boosting algorithm for multimodal magnetic resonance imaging-based infarct prediction. METHODS: In a retrospective analysis of 195 patients with acute ischemic stroke, fluid-attenuated inversion recovery, diffusion-weighted imaging, and 10 perfusion parameters were derived from acute magnetic resonance imaging scans. They were integrated to predict final infarct as seen on follow-up T2-fluid-attenuated inversion recovery using the extreme gradient boosting and compared with a standard generalized linear model approach using cross-validation. Submodels for recanalization and persistent occlusion were calculated and were used to identify the important imaging markers. Performance in infarct prediction was analyzed with receiver operating characteristics. Resulting areas under the curve and accuracy rates were compared using Wilcoxon signed-rank test. RESULTS: The extreme gradient boosting model demonstrated significantly higher performance in infarct prediction compared with generalized linear model in both cross-validation approaches: 5-folds (P<10e-16) and leave-one-out (P<0.015). The imaging parameters time-to-peak, mean transit time, time-to-maximum, and diffusion-weighted imaging were indicated as most valuable for infarct prediction by the systematic algorithm rating. Notably, the performance improvement was higher with 5-folds cross-validation approach than leave-one-out. CONCLUSIONS: We demonstrate extreme gradient boosting as a state-of-the-art method for clinically applicable multimodal magnetic resonance imaging infarct prediction in acute ischemic stroke. Our findings emphasize the role of perfusion parameters as important biomarkers for infarct prediction. The effect of cross-validation techniques on performance indicates that the intrapatient variability is expressed in nonlinear dynamics of the imaging modalities.

The effects of hypercapnia on cortical capillary transit time heterogeneity (CTH) in anesthetized mice.

Capillary flow patterns are highly heterogeneous in the resting brain. During hyperemia, capillary transit-time heterogeneity (CTH) decreases, in proportion to blood's mean transit time (MTT) in passive, compliant microvascular networks. Previously, we found that functional activation reduces the CTH:MTT ratio, suggesting that additional homogenization takes place through active neurocapillary coupling mechanisms. Here, we examine changes in the CTH:MTT ratio during hypercapnic hyperemia in anesthetized mice (C57Bl/6NTac), expecting that homogenization is smaller than during functional hyperemia. We used an indicator-dilution technique and multiple capillary scans by two-photon microscopy to estimate CTH and MTT. During hypercapnia, MTT and CTH decreased as derived from indicator-dilution between artery and vein, as well as between arterioles and venules. The CTH:MTT ratio, however, increased. The same tendency was observed in the estimates from capillary scans. The parallel reductions of MTT and CTH are consistent with previous data. We speculate that the relative increase in CTH compared to MTT during hypercapnia represents either or both capillary constrictions and blood passage through functional thoroughfare channels. Intriguingly, hemodynamic responses to hypercapnia declined with cortical depth, opposite previous reports of hemodynamic responses to functional activation. Our findings support the role of CTH in cerebral flow-metabolism coupling during hyperemia.

Recurrence of glioblastoma is associated with elevated microvascular transit time heterogeneity and increased hypoxia.

Dynamic susceptibility contrast (DSC) perfusion MRI provide information about differences in macro- and microvasculature when executed with gradient-echo (GE; sensitive to macrovasculature) and spin-echo (SE; sensitive to microvasculature) contrast. This study investigated whether there are differences between macro- and microvascular transit time heterogeneity (MVTH and µVTH) and tissue oxygen tension (PO2mit) in newly-diagnosed and recurrent glioblastoma. Fifty-seven patients with glioblastoma (25 newly-diagnosed/32 recurrent) were examined with GE- and SE-DSC perfusion sequences, and a quantitative blood-oxygen-level-dependent (qBOLD) approach. Maps of MVTH, µVTH and coefficient of variation (MCOV and µCOV) were calculated from GE- and SE-DSC data, respectively, using an extended flow-diffusion equation. PO2mit maps were calculated from qBOLD data. Newly-diagnosed and recurrent glioblastoma showed significantly lower ( P ≤ 0.001) µCOV values compared to both normal brain and macrovasculature (MCOV) of the lesions. Recurrent glioblastoma had significantly higher µVTH ( P = 0.014) and µCOV ( P = 0.039) as well as significantly lower PO2mit values ( P = 0.008) compared to newly-diagnosed glioblastoma. The macrovasculature, however, showed no significant differences. Our findings provide evidence of microvascular adaption in the disorganized tumor vasculature for retaining the metabolic demands in stress response of therapeutically-uncontrolled glioblastomas. Thus, µVTH and PO2mit mapping gives insight into the tumor microenvironment (vascular and hypoxic niches) responsible for therapy resistance.

Transit time homogenization in ischemic stroke - A novel biomarker of penumbral microvascular failure?

Cerebral ischemia causes widespread capillary no-flow in animal studies. The extent of microvascular impairment in human stroke, however, is unclear. We examined how acute intra-voxel transit time characteristics and subsequent recanalization affect tissue outcome on follow-up MRI in a historic cohort of 126 acute ischemic stroke patients. Based on perfusion-weighted MRI data, we characterized voxel-wise transit times in terms of their mean transit time (MTT), standard deviation (capillary transit time heterogeneity - CTH), and the CTH:MTT ratio (relative transit time heterogeneity), which is expected to remain constant during changes in perfusion pressure in a microvasculature consisting of passive, compliant vessels. To aid data interpretation, we also developed a computational model that relates graded microvascular failure to changes in these parameters. In perfusion-diffusion mismatch tissue, prolonged mean transit time (>5 seconds) and very low cerebral blood flow (≤6 mL/100 mL/min) was associated with high risk of infarction, largely independent of recanalization status. In the remaining mismatch region, low relative transit time heterogeneity predicted subsequent infarction if recanalization was not achieved. Our model suggested that transit time homogenization represents capillary no-flow. Consistent with this notion, low relative transit time heterogeneity values were associated with lower cerebral blood volume. We speculate that low RTH may represent a novel biomarker of penumbral microvascular failure.

ABCD1 dysfunction alters white matter microvascular perfusion.

Cerebral X-linked adrenoleukodystrophy is a devastating neurodegenerative disorder caused by mutations in the ABCD1 gene, which lead to a rapidly progressive cerebral inflammatory demyelination in up to 60% of affected males. Selective brain endothelial dysfunction and increased permeability of the blood-brain barrier suggest that white matter microvascular dysfunction contributes to the conversion to cerebral disease. Applying a vascular model to conventional dynamic susceptibility contrast magnetic resonance perfusion imaging, we demonstrate that lack of ABCD1 function causes increased capillary flow heterogeneity in asymptomatic hemizygotes predominantly in the white matter regions and developmental stages with the highest probability for conversion to cerebral disease. In subjects with ongoing inflammatory demyelination we observed a sequence of increased capillary flow heterogeneity followed by blood-brain barrier permeability changes in the perilesional white matter, which predicts lesion progression. These white matter microvascular alterations normalize within 1 year after treatment with haematopoietic stem cell transplantation. For the first time in vivo, our studies unveil a model to assess how ABCD1 alters white matter microvascular function and explores its potential as an earlier biomarker for monitoring disease progression and response to treatment.