Infarcts Due to Large Vessel Occlusions Continue to Grow Despite Near-Complete Reperfusion After Endovascular Treatment.

BACKGROUND AND PURPOSE: Infarcts in acute ischemic stroke (AIS) patients may continue to grow even after reperfusion, due to mechanisms such as microvascular obstruction and reperfusion injury. We investigated whether and how much infarcts grow in AIS patients after near-complete (expanded Thrombolysis in Cerebral Infarction [eTICI] 2c/3) reperfusion following endovascular treatment (EVT), and to assess the association of post-reperfusion infarct growth with clinical outcomes. METHODS: Data are from a single-center retrospective observational cohort study that included AIS patients undergoing EVT with near-complete reperfusion who received diffusion-weighted magnetic resonance imaging (MRI) within 2 hours post-EVT and 24 hours after EVT. Association of infarct growth between 2 and 24 hours post-EVT and 24-hour National Institutes of Health Stroke Scale (NIHSS) as well as 90-day modified Rankin Scale score was assessed using multivariable logistic regression. RESULTS: Ninety-four of 155 (60.6%) patients achieved eTICI 2c/3 and were included in the analysis. Eighty of these 94 (85.1%) patients showed infarct growth between 2 and 24 hours post-reperfusion. Infarct growth ≥5 mL was seen in 39/94 (41.5%) patients, and infarct growth ≥10 mL was seen in 20/94 (21.3%) patients. Median infarct growth between 2 and 24 hours post-reperfusion was 4.5 mL (interquartile range: 0.4-9.2 mL). Post-reperfusion infarct growth was associated with the 24-hour NIHSS in multivariable analysis (odds ratio: 1.16 [95% confidence interval 1.09-1.24], P<0.01). CONCLUSION: Infarcts continue to grow after EVT, even if near-complete reperfusion is achieved. Investigating the underlying mechanisms may inform future therapeutic approaches for mitigating the process and help improve patient outcome.

MRI-Based Prediction of Clinical Improvement Following Ventricular Shunt Placement for Normal Pressure Hydrocephalus (NPH): Development and Evaluation of an Integrated Multi-Sequence Machine Learning Algorithm.

BACKGROUND AND PURPOSE: Symptoms of normal pressure hydrocephalus (NPH) are sometimes refractory to shunt placement, with limited ability to predict improvement for individual patients. We evaluated an MRI-based artificial intelligence method to predict post-shunt NPH symptom improvement. MATERIALS AND METHODS: NPH patients who underwent magnetic resonance imaging (MRI) prior to shunt placement at a single center (2014-2021) were identified. Twelve-month post-shunt improvement in modified Rankin Scale (mRS), incontinence, gait, and cognition were retrospectively abstracted from clinical documentation. 3D deep residual neural networks were built on skull stripped T2-weighted and fluid attenuated inversion recovery (FLAIR) images. Predictions based on both sequences were fused by additional network layers. Patients from 2014-2019 were used for parameter optimization, while those from 2020-2021 were used for testing. Models were validated on an external validation dataset from a second institution (n=33). RESULTS: Of 249 patients, n=201 and n=185 were included in the T2-based and FLAIR-based models according to imaging availability. The combination of T2-weighted and FLAIR sequences offered the best performance in mRS and gait improvement predictions relative to models trained on imaging acquired using only one sequence, with AUROC values of 0.7395 [0.5765-0.9024] for mRS and 0.8816 [0.8030-0.9602] for gait. For urinary incontinence and cognition, combined model performances on predicting outcomes were similar to FLAIR-only performance, with AUROC values of 0.7874 [0.6845-0.8903] and 0.7230 [0.5600-0.8859]. CONCLUSIONS: Application of a combined algorithm using both T2-weighted and FLAIR sequences offered the best image-based prediction of post-shunt symptom improvement, particularly for gait and overall function in terms of mRS. ABBREVIATIONS: NPH = normal pressure hydrocephalus; iNPH = idiopathic NPH; sNPH = secondary NPH; AI = artificial intelligence; ML = machine learning; CSF = cerebrospinal fluid; AUROC = area under the receiver operating characteristic; FLAIR = fluid attenuated inversion recovery; BMI = body mass index; CCI = Charlson Comorbidity Index; SD = standard deviation; IQR = interquartile range.

Identification of a single-dose, low flip angle based CBV threshold for fractional tumor burden (FTB) mapping in recurrent glioblastoma.

BACKGROUND AND PURPOSE: DSC-MRI can be used to generate fractional tumor burden (FTB) maps, via application of relative CBV thresholds, to spatially differentiate glioblastoma recurrence from post treatment radiation effects (PTRE). Image-localized histopathology was previously used to validate FTB maps derived from a reference DSC-MRI protocol using preload, a moderate flip angle (MFA, 60°) and post-processing leakage correction. Recently, a DSC-MRI protocol with a low flip angle (LFA, 30°) with no preload was shown to provide leakage-corrected RCBV equivalent to the reference protocol. This study aims to identify the RCBV thresholds for the LFA protocol that generate the most accurate FTB maps, concordant with those obtained from the reference MFA protocol. MATERIALS AND METHODS: Fifty-two patients with grade IV GBM who had prior surgical resection and received chemotherapy and radiotherapy were included in the study. Two sets of DSC-MRI data were collected sequentially first using LFA protocol with no preload, which served as the preload for the subsequent MFA protocol. Standardized relative CBV maps (sRCBV) were obtained for each patient and co-registered with the anatomical post-contrast T1-weighted images. The reference MFA-based FTB maps were computed using previously published sRCBV thresholds (1.0 and 1.56). An ROC analysis was conducted to identify the optimal, voxelwise LFA sRCBV thresholds, and the sensitivity, specificity, and accuracy of the LFA-based FTB maps were computed with respect to the MFA-based reference. RESULTS: The mean sRCBV values of tumors across patients exhibited strong agreement (CCC = 0.99) between the two protocols. Using the ROC analysis, the optimal lower LFA threshold that accurately distinguishes PTRE from tumor recurrence was found to be 1.0 (sensitivity: 87.77%; specificity: 90.22%), equivalent to the ground truth. To identify aggressive tumor regions, the ROC analysis identified an upper LFA threshold of 1.37 (sensitivity: 90.87%; specificity: 91.10%) for the reference MFA threshold of 1.56. CONCLUSION: For LFA-based FTB maps, a sRCBV threshold of 1.0 and 1.37 can differentiate PTRE from recurrent tumor. FTB maps aids in surgical planning, guiding pathological diagnosis and treatment strategies in the recurrent setting. This study further confirms the reliability of single-dose LFA-based DSC-MRI. ABBREVIATIONS: LFA = low flip angle; MFA = moderate flip angle; sRCBV = standardized relative cerebral blood volume; FTB = fractional tumor burden; PTRE = post treatment radiation effects; ROC = receiver operating characteristics; CCC = concordance correlation coefficient.

Modeling diffusion-weighted imaging lesion expansion between 2 and 24 h after endovascular thrombectomy in acute ischemic stroke.

PURPOSE: Diffusion-weighted imaging (DWI) lesion expansion after endovascular thrombectomy (EVT) is not well characterized. We used serial diffusion-weighted magnetic resonance imaging (MRI) to measure lesion expansion between 2 and 24 h after EVT. METHODS: In this single-center observational analysis of patients with acute ischemic stroke due to large vessel occlusion, DWI was performed post-EVT (< 2 h after closure) and 24-h later. DWI lesion expansion was evaluated using multivariate generalized linear mixed modeling with various clinical moderators. RESULTS: We included 151 patients, of which 133 (88%) had DWI lesion expansion, defined as a positive change in lesion volume between 2 and 24 h. In an unadjusted analysis, median baseline DWI lesion volume immediately post-EVT was 15.0 mL (IQR: 6.6-36.8) and median DWI lesion volume 24 h post-EVT was 20.8 mL (IQR: 9.4-66.6), representing a median change of 6.1 mL (IQR: 1.5-17.7), or a 39% increase. There were no significant associations among univariable models of lesion expansion. Adjusted models of DWI lesion expansion demonstrated that relative lesion expansion (defined as final/initial DWI lesion volume) was consistent across eTICI scores (0-2a, 0.52%; 2b, 0.49%; 2c-3, 0.42%, p = 0.69). For every 1 mL increase in lesion volume, there was 2% odds of an increase in 90-day mRS (OR: 1.021, 95%CI [1.009, 1.034], p < 0.001). CONCLUSION: We observed substantial lesion expansion post-EVT whereby relative lesion expansion was consistent across eTICI categories, and greater absolute lesion expansion was associated with worse clinical outcome. Our findings suggest that alternate endpoints for cerebroprotectant trials may be feasible.

A cross-sectional study to test equivalence of low- versus intermediate-flip angle dynamic susceptibility contrast MRI measures of relative cerebral blood volume in patients with high-grade gliomas at 1.5 Tesla field strength.

Introduction: 1.5 Tesla (1.5T) remain a significant field strength for brain imaging worldwide. Recent computer simulations and clinical studies at 3T MRI have suggested that dynamic susceptibility contrast (DSC) MRI using a 30° flip angle ("low-FA") with model-based leakage correction and no gadolinium-based contrast agent (GBCA) preload provides equivalent relative cerebral blood volume (rCBV) measurements to the reference-standard acquisition using a single-dose GBCA preload with a 60° flip angle ("intermediate-FA") and model-based leakage correction. However, it remains unclear whether this holds true at 1.5T. The purpose of this study was to test this at 1.5T in human high-grade glioma (HGG) patients. Methods: This was a single-institution cross-sectional study of patients who had undergone 1.5T MRI for HGG. DSC-MRI consisted of gradient-echo echo-planar imaging (GRE-EPI) with a low-FA without preload (30°/P-); this then subsequently served as a preload for the standard intermediate-FA acquisition (60°/P+). Both normalized (nrCBV) and standardized relative cerebral blood volumes (srCBV) were calculated using model-based leakage correction (C+) with IBNeuro™ software. Whole-enhancing lesion mean and median nrCBV and srCBV from the low- and intermediate-FA methods were compared using the Pearson's, Spearman's and intraclass correlation coefficients (ICC). Results: Twenty-three HGG patients composing a total of 31 scans were analyzed. The Pearson and Spearman correlations and ICCs between the 30°/P-/C+ and 60°/P+/C+ acquisitions demonstrated high correlations for both mean and median nrCBV and srCBV. Conclusion: Our study provides preliminary evidence that for HGG patients at 1.5T MRI, a low FA, no preload DSC-MRI acquisition can be an appealing alternative to the reference standard higher FA acquisition that utilizes a preload.

Early post-bevacizumab change in rCBV from DSC-MRI identifies pseudoresponse in recurrent glioblastoma: Results from ACRIN 6677/RTOG 0625.

Background: Progressive enhancement predicted poor survival in ACRIN 6677/RTOG 0625, a multi-center trial of bevacizumab with irinotecan or temozolomide in recurrent glioblastoma, but pseudoresponse likely limited enhancement-based survival prognostication in T1 non-progressors. We aimed to determine whether early change in cerebral blood volume from baseline (ΔCBV) could further stratify the T1 non-progressors according to overall (OS) and progression-free (PFS) survival. Methods: 37/123 enrolled patients had DSC-MRI, including 13, 15, and 8 patients without 2D-T1 progression at 2, 8, and 16 weeks post-treatment initiation, respectively. Mean CBV normalized to white matter (nRCBV) and mean standardized CBV (sRCBV) were extracted from enhancing tumor. ROC curves were derived for ΔCBV using six-month PFS and one-year OS as reference standards. Kaplan-Meier survival estimates and log-rank test compared PFS and OS for both ΔCBV (increase vs. decrease) and T1 response status (stable vs. decreasing enhancement). Results: PFS and OS were significantly worse for increasing CBV at 2 weeks (p=0.003 and p=0.002 for nRCBV, and p=0.03 and p=0.03 for sRCBV, respectively), but not for 2D-T1 patients with stable vs. decreasing enhancement (p=0.44 and p=0.86, respectively). ΔCBV at week 2 was also a good prognostic marker for OS-1 and PFS-6 using ROC analysis. By contrast, 2D-T1 response status at weeks 2, 8, and 16 was not associated with PFS-6. ΔCBV at 16 weeks (p=0.008 for sRCBV) but not 8 weeks (p=0.74 for nRCBV and p=0.56 for sRCBV) was associated with significant difference in median survival, but no difference in survival was observed for 2D-T1 patients with stable vs. decreasing enhancement at 8 weeks (p=0.69) or 16 weeks (p=0.21). At 16 weeks, OS did not differ significantly between 2D-T1 progressors and 2D-T1 non-progressors with increasing CBV (median survival 3.3 months post week 16 scan vs. 9.2 months, respectively; p=0.13), suggesting that 2D-T1 non-progressors with increasing CBV may have a prognosis like that of 2D-T1 progressors. Conclusion: After 2 weeks of anti-angiogenic therapy, ΔCBV in 2D-T1 non-progressors significantly prognosticated PFS and OS, whereas 2D-T1 response status did not, identifying a subpopulation that benefits from bevacizumab. Combining 2D-T1 progression and ΔCBV may yield a response assessment paradigm with 3-tiered OS stratification.

Hypothetical generalized framework for a new imaging endpoint of therapeutic activity in early phase clinical trials in brain tumors.

Imaging response assessment is a cornerstone of patient care and drug development in oncology. Clinicians/clinical researchers rely on tumor imaging to estimate the impact of new treatments and guide decision making for patients and candidate therapies. This is important in brain cancer, where associations between tumor size/growth and emerging neurological deficits are strong. Accurately measuring the impact of a new therapy on tumor growth early in clinical development, where patient numbers are small, would be valuable for decision making regarding late-stage development activation. Current attempts to measure the impact of a new therapy have limited influence on clinical development, as determination of progression, stability or response does not currently account for individual tumor growth kinetics prior to the initiation of experimental therapies. Therefore, we posit that imaging-based response assessment, often used as a tool for estimating clinical effect, is incomplete as it does not adequately account for growth trajectories or biological characteristics of tumors prior to the introduction of an investigational agent. Here, we propose modifications to the existing framework for evaluating imaging assessment in primary brain tumors that will provide a more reliable understanding of treatment effects. Measuring tumor growth trajectories prior to a given intervention may allow us to more confidently conclude whether there is an anti-tumor effect. This updated approach to imaging-based tumor response assessment is intended to improve our ability to select candidate therapies for later-stage development, including those that may not meet currently sought thresholds for "response" and ultimately lead to identification of effective treatments.

Deep learning-based automatic tumor burden assessment of pediatric high-grade gliomas, medulloblastomas, and other leptomeningeal seeding tumors.

BACKGROUND: Longitudinal measurement of tumor burden with magnetic resonance imaging (MRI) is an essential component of response assessment in pediatric brain tumors. We developed a fully automated pipeline for the segmentation of tumors in pediatric high-grade gliomas, medulloblastomas, and leptomeningeal seeding tumors. We further developed an algorithm for automatic 2D and volumetric size measurement of tumors. METHODS: The preoperative and postoperative cohorts were randomly split into training and testing sets in a 4:1 ratio. A 3D U-Net neural network was trained to automatically segment the tumor on T1 contrast-enhanced and T2/FLAIR images. The product of the maximum bidimensional diameters according to the RAPNO (Response Assessment in Pediatric Neuro-Oncology) criteria (AutoRAPNO) was determined. Performance was compared to that of 2 expert human raters who performed assessments independently. Volumetric measurements of predicted and expert segmentations were computationally derived and compared. RESULTS: A total of 794 preoperative MRIs from 794 patients and 1003 postoperative MRIs from 122 patients were included. There was excellent agreement of volumes between preoperative and postoperative predicted and manual segmentations, with intraclass correlation coefficients (ICCs) of 0.912 and 0.960 for the 2 preoperative and 0.947 and 0.896 for the 2 postoperative models. There was high agreement between AutoRAPNO scores on predicted segmentations and manually calculated scores based on manual segmentations (Rater 2 ICC = 0.909; Rater 3 ICC = 0.851). Lastly, the performance of AutoRAPNO was superior in repeatability to that of human raters for MRIs with multiple lesions. CONCLUSIONS: Our automated deep learning pipeline demonstrates potential utility for response assessment in pediatric brain tumors. The tool should be further validated in prospective studies.

Advanced MRI Protocols to Discriminate Glioma From Treatment Effects: State of the Art and Future Directions.

In the follow-up treatment of high-grade gliomas (HGGs), differentiating true tumor progression from treatment-related effects, such as pseudoprogression and radiation necrosis, presents an ongoing clinical challenge. Conventional MRI with and without intravenous contrast serves as the clinical benchmark for the posttreatment surveillance imaging of HGG. However, many advanced imaging techniques have shown promise in helping better delineate the findings in indeterminate scenarios, as posttreatment effects can often mimic true tumor progression on conventional imaging. These challenges are further confounded by the histologic admixture that can commonly occur between tumor growth and treatment-related effects within the posttreatment bed. This review discusses the current practices in the surveillance imaging of HGG and the role of advanced imaging techniques, including perfusion MRI and metabolic MRI.

Encephalopathy at admission predicts adverse outcomes in patients with SARS-CoV-2 infection.

AIMS: To determine if neurologic symptoms at admission can predict adverse outcomes in patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). METHODS: Electronic medical records of 1053 consecutively hospitalized patients with laboratory-confirmed infection of SARS-CoV-2 from one large medical center in the USA were retrospectively analyzed. Univariable and multivariable Cox regression analyses were performed with the calculation of areas under the curve (AUC) and concordance index (C-index). Patients were stratified into subgroups based on the presence of encephalopathy and its severity using survival statistics. In sensitivity analyses, patients with mild/moderate and severe encephalopathy (defined as coma) were separately considered. RESULTS: Of 1053 patients (mean age 52.4 years, 48.0% men [n = 505]), 35.1% (n = 370) had neurologic manifestations at admission, including 10.3% (n = 108) with encephalopathy. Encephalopathy was an independent predictor for death (hazard ratio [HR] 2.617, 95% confidence interval [CI] 1.481-4.625) in multivariable Cox regression. The addition of encephalopathy to multivariable models comprising other predictors for adverse outcomes increased AUCs (mortality: 0.84-0.86, ventilation/ intensive care unit [ICU]: 0.76-0.78) and C-index (mortality: 0.78 to 0.81, ventilation/ICU: 0.85-0.86). In sensitivity analyses, risk stratification survival curves for mortality and ventilation/ICU based on severe encephalopathy (n = 15) versus mild/moderate encephalopathy (n = 93) versus no encephalopathy (n = 945) at admission were discriminative (p < 0.001). CONCLUSIONS: Encephalopathy at admission predicts later progression to death in SARS-CoV-2 infection, which may have important implications for risk stratification in clinical practice.

The relationship between cerebral and retinal microbleeds in cerebral amyloid angiopathy (CAA): A pilot study.

BACKGROUND: The standard in vivo diagnostic imaging technique for cerebral amyloid angiopathy (CAA) is costly and thereby of limited utility for point-of-care diagnosis and monitoring of treatment efficacy. Recent recognition that retinal changes may reflect cerebral changes in neurodegenerative disease provides an ideal opportunity for development of accessible and cost-effective biomarkers for point-of-care use in the detection and monitoring of CAA. In this pilot study, we examined structural and angiographic retinal changes in CAA patients relative to a control group, and compared retinal and cerebral pathology in a group of CAA patients. METHODS: We used spectral domain optical coherence tomography (SD-OCT) to image the retina and compared retinal microbleeds to both cerebral microbleeds and white matter hyperintensities (WMH) in CAA patients, as seen on MRI. We compared retinal angiographic changes, along with structural retinal neuronal layer changes in CAA patients and cognitively normal older adults, and examined the relationship between retinal and cerebral microbleeds and cognition in CAA patients. RESULTS: We found a trend level correlation between retinal and cerebral microbleeds in CAA patients. Moreover, we found a significant correlation between retinal microbleeds and episodic memory performance in CAA patients. There were no significant group differences between CAA patients and cognitively normal older adults on retinal angiographic or structural measurements. CONCLUSION: Retinal microbleeds may reflect degree of cerebral microbleed burden in CAA. This picture was complicated by systolic hypertension in the CAA group, which is a confounding factor for the interpretation of these data. Our results stimulate motivation for pursuit of a more comprehensive prospective study to determine the feasibility of retinal biomarkers in CAA.

Consensus recommendations for MRI and PET imaging of primary central nervous system lymphoma: guideline statement from the International Primary CNS Lymphoma Collaborative Group (IPCG).

Advanced molecular and pathophysiologic characterization of primary central nervous system lymphoma (PCNSL) has revealed insights into promising targeted therapeutic approaches. Medical imaging plays a fundamental role in PCNSL diagnosis, staging, and response assessment. Institutional imaging variation and inconsistent clinical trial reporting diminishes the reliability and reproducibility of clinical response assessment. In this context, we aimed to: (1) critically review the use of advanced positron emission tomography (PET) and magnetic resonance imaging (MRI) in the setting of PCNSL; (2) provide results from an international survey of clinical sites describing the current practices for routine and advanced imaging, and (3) provide biologically based recommendations from the International PCNSL Collaborative Group (IPCG) on adaptation of standardized imaging practices. The IPCG provides PET and MRI consensus recommendations built upon previous recommendations for standardized brain tumor imaging protocols (BTIP) in primary and metastatic disease. A biologically integrated approach is provided to addresses the unique challenges associated with the imaging assessment of PCNSL. Detailed imaging parameters facilitate the adoption of these recommendations by researchers and clinicians. To enhance clinical feasibility, we have developed both "ideal" and "minimum standard" protocols at 3T and 1.5T MR systems that will facilitate widespread adoption.

Robust breast cancer detection in mammography and digital breast tomosynthesis using an annotation-efficient deep learning approach.

Breast cancer remains a global challenge, causing over 600,000 deaths in 2018 (ref. 1). To achieve earlier cancer detection, health organizations worldwide recommend screening mammography, which is estimated to decrease breast cancer mortality by 20-40% (refs. 2,3). Despite the clear value of screening mammography, significant false positive and false negative rates along with non-uniformities in expert reader availability leave opportunities for improving quality and access4,5. To address these limitations, there has been much recent interest in applying deep learning to mammography6-18, and these efforts have highlighted two key difficulties: obtaining large amounts of annotated training data and ensuring generalization across populations, acquisition equipment and modalities. Here we present an annotation-efficient deep learning approach that (1) achieves state-of-the-art performance in mammogram classification, (2) successfully extends to digital breast tomosynthesis (DBT; '3D mammography'), (3) detects cancers in clinically negative prior mammograms of patients with cancer, (4) generalizes well to a population with low screening rates and (5) outperforms five out of five full-time breast-imaging specialists with an average increase in sensitivity of 14%. By creating new 'maximum suspicion projection' (MSP) images from DBT data, our progressively trained, multiple-instance learning approach effectively trains on DBT exams using only breast-level labels while maintaining localization-based interpretability. Altogether, our results demonstrate promise towards software that can improve the accuracy of and access to screening mammography worldwide.

Value of dynamic contrast perfusion MRI to predict early response to bevacizumab in newly diagnosed glioblastoma: results from ACRIN 6686 multicenter trial.

BACKGROUND: In Radiation Therapy Oncology Group (RTOG) 0825, a phase III trial of standard therapy with bevacizumab or without (placebo) in newly diagnosed glioblastoma, 44 patients underwent dynamic contrast enhanced (DCE) and/or dynamic susceptibility contrast (DSC) MRI in the American College of Radiology Imaging Network (ACRIN) trial 6686. The association between early changes in relative cerebral blood volume (rCBV) and volume transfer constant (Ktrans) with overall survival (OS) was evaluated. METHODS: MRI was performed at postop baseline (S0), immediately before (S1), 1 day after (S2), and 7 weeks after (S3) bevacizumab or placebo initiation. Mean normalized and standardized rCBV (nRCBV, sRCBV) and Ktrans were measured within contrast-enhancing lesion. Wilcoxon rank sum tests compared parameter changes from S1-S2 and S1-S3. Association with OS and progression-free survival (PFS) were determined using Kaplan-Meier and log-rank tests. Treatment response for groups stratified by pretreatment nRCBV (S0, S1) was explored. The intraclass correlation coefficient and repeatability coefficient for the placebo arm (S1-S2) were used to assess repeatability. RESULTS: Evaluable were 27-36 datasets per time point. Significant differences between treatment arms were found for changes in nRCBV and sRCBV from S1-S2 and S1-S3, and in Ktrans for S1-S3. Improved PFS (P = 0.05) but not OS (P = 0.46) was observed. High pretreatment rCBV predicted improved OS for bevacizumab-treated patients. Based on the intraclass correlation coefficient, sRCBV (0.92) was more repeatable than nRCBV (0.71) and Ktrans (0.75), consistent with repeatability coefficient values. CONCLUSIONS: Bevacizumab significantly changes rCBV but not Ktrans as early as 1 day posttreatment in newly diagnosed glioblastoma unrelated to outcomes. Improvements in clinical trial design to maximize rCBV benefit are indicated.

Radiographic read paradigms and the roles of the central imaging laboratory in neuro-oncology clinical trials.

Determination of therapeutic benefit in intracranial tumors is intimately dependent on serial assessment of radiographic images. The Response Assessment in Neuro-Oncology (RANO) criteria were established in 2010 to provide an updated framework to better characterize tumor response to contemporary treatments. Since this initial update a number of RANO criteria have provided some basic principles for the interpretation of changes on MR images; however, the details of how to operationalize RANO and other criteria for use in clinical trials are ambiguous and not standardized. In this review article designed for the neuro-oncologist or treating clinician, we outline essential steps for performing radiographic assessments by highlighting primary features of the Imaging Charter (referred to as the Charter for the remainder of this article), a document that describes the clinical trial imaging methodology and methods to ensure operationalization of the Charter into the workings of a clinical trial. Lastly, we provide recommendations for specific changes to optimize this methodology for neuro-oncology, including image registration, requirement of growing tumor for eligibility in trials of recurrent tumor, standardized image acquisition guidelines, and hybrid reader paradigms that allow for both unbiased measurements and more comprehensive interpretation.

Deep neural network to locate and segment brain tumors outperformed the expert technicians who created the training data.

Purpose: Deep learning (DL) algorithms have shown promising results for brain tumor segmentation in MRI. However, validation is required prior to routine clinical use. We report the first randomized and blinded comparison of DL and trained technician segmentations. Approach: We compiled a multi-institutional database of 741 pretreatment MRI exams. Each contained a postcontrast T1-weighted exam, a T2-weighted fluid-attenuated inversion recovery exam, and at least one technician-derived tumor segmentation. The database included 729 unique patients (470 males and 259 females). Of these exams, 641 were used for training the DL system, and 100 were reserved for testing. We developed a platform to enable qualitative, blinded, controlled assessment of lesion segmentations made by technicians and the DL method. On this platform, 20 neuroradiologists performed 400 side-by-side comparisons of segmentations on 100 test cases. They scored each segmentation between 0 (poor) and 10 (perfect). Agreement between segmentations from technicians and the DL method was also evaluated quantitatively using the Dice coefficient, which produces values between 0 (no overlap) and 1 (perfect overlap). Results: The neuroradiologists gave technician and DL segmentations mean scores of 6.97 and 7.31, respectively ( p < 0.00007 ). The DL method achieved a mean Dice coefficient of 0.87 on the test cases. Conclusions: This was the first objective comparison of automated and human segmentation using a blinded controlled assessment study. Our DL system learned to outperform its "human teachers" and produced output that was better, on average, than its training data.

Detecting Large Vessel Occlusion at Multiphase CT Angiography by Using a Deep Convolutional Neural Network.

Background Large vessel occlusion (LVO) stroke is one of the most time-sensitive diagnoses in medicine and requires emergent endovascular therapy to reduce morbidity and mortality. Leveraging recent advances in deep learning may facilitate rapid detection and reduce time to treatment. Purpose To develop a convolutional neural network to detect LVOs at multiphase CT angiography. Materials and Methods This multicenter retrospective study evaluated 540 adults with CT angiography examinations for suspected acute ischemic stroke from February 2017 to June 2018. Examinations positive for LVO (n = 270) were confirmed by catheter angiography and LVO-negative examinations (n = 270) were confirmed through review of clinical and radiology reports. Preprocessing of the CT angiography examinations included vasculature segmentation and the creation of maximum intensity projection images to emphasize the contrast agent-enhanced vasculature. Seven experiments were performed by using combinations of the three phases (arterial, phase 1; peak venous, phase 2; and late venous, phase 3) of the CT angiography. Model performance was evaluated on the held-out test set. Metrics included area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. Results The test set included 62 patients (mean age, 69.5 years; 48% women). Single-phase CT angiography achieved an AUC of 0.74 (95% confidence interval [CI]: 0.63, 0.85) with sensitivity of 77% (24 of 31; 95% CI: 59%, 89%) and specificity of 71% (22 of 31; 95% CI: 53%, 84%). Phases 1, 2, and 3 together achieved an AUC of 0.89 (95% CI: 0.81, 0.96), sensitivity of 100% (31 of 31; 95% CI: 99%, 100%), and specificity of 77% (24 of 31; 95% CI: 59%, 89%), a statistically significant improvement relative to single-phase CT angiography (P = .01). Likewise, phases 1 and 3 and phases 2 and 3 also demonstrated improved fit relative to single phase (P = .03). Conclusion This deep learning model was able to detect the presence of large vessel occlusion and its diagnostic performance was enhanced by using delayed phases at multiphase CT angiography examinations. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Ospel and Goyal in this issue.