Thrombus Attenuation Gradient Can Predict Successful First-Pass Recanalization Following Stentriever Thrombectomy.

OBJECTIVE: Computed tomography angiography (CTA) derived thrombus enhancement characteristics can predict first-pass recanalization. We studied whether dynamic contrast kinetics within the clot in multiphase CTA can predict first-pass recanalization following stentriever thrombectomy. METHODS: Patients with acute large vessel occlusive stroke evaluated with multiphasic CTA who underwent stentriever thrombectomy were selected. Thrombus perviousness on various phases including arterial, venous, and delayed phases was calculated. Thrombus attenuation gradient (TAG), defined as average attenuation difference between adjacent phases, was also evaluated and correlated with successful first-pass outcome (modified Treatment in Cerebral Ischemia score ≥2b). RESULTS: Of 69 patients, 32 (47%) had successful first-pass recanalization (group 1), and 37 (53%) required >1 attempt (group 2). TAG showed significant differences in arterial-plain and venous-arterial phases. The early increase in TAG was seen in group 1 in the arterial-plain phase, as opposed to group 2 (12.6 vs. 9, P = 0.01), which plateaued in the venous-arterial phase for group 1 and showed a further increase in group 2 (2.1 vs. 5.1, P = 0.02). A cutoff value of 9.2 HU for arterial-plain phase (P = 0.001) and 4.2 HU (P = 0.001) for venous-arterial phase was predictive of first-pass effect. Combining 2 metrics had an odds ratio of 2.8 for first-pass recanalization (P = 0.035). Accuracy evaluated in a validation cohort yielded 74%. Other features including histology were not significant. CONCLUSIONS: TAG evaluated from multiphase CTA can predict first-pass effect in stentriever thrombectomy.

Region of interest focused MRI to synthetic CT translation using regression and segmentation multi-task network.

Objective. In MR-only clinical workflow, replacing CT with MR image is of advantage for workflow efficiency and reduces radiation to the patient. An important step required to eliminate CT scan from the workflow is to generate the information provided by CT via an MR image. In this work, we aim to demonstrate a method to generate accurate synthetic CT (sCT) from an MR image to suit the radiation therapy (RT) treatment planning workflow. We show the feasibility of the method and make way for a broader clinical evaluation.Approach. We present a machine learning method for sCT generation from zero-echo-time (ZTE) MRI aimed at structural and quantitative accuracies of the image, with a particular focus on the accurate bone density value prediction. The misestimation of bone density in the radiation path could lead to unintended dose delivery to the target volume and results in suboptimal treatment outcome. We propose a loss function that favors a spatially sparse bone region in the image. We harness the ability of the multi-task network to produce correlated outputs as a framework to enable localization of region of interest (RoI) via segmentation, emphasize regression of values within RoI and still retain the overall accuracy via global regression. The network is optimized by a composite loss function that combines a dedicated loss from each task.Main results. We have included 54 brain patient images in this study and tested the sCT images against reference CT on a subset of 20 cases. A pilot dose evaluation was performed on 9 of the 20 test cases to demonstrate the viability of the generated sCT in RT planning. The average quantitative metrics produced by the proposed method over the test set were-(a) mean absolute error (MAE) of 70 ± 8.6 HU; (b) peak signal-to-noise ratio (PSNR) of 29.4 ± 2.8 dB; structural similarity metric (SSIM) of 0.95 ± 0.02; and (d) Dice coefficient of the body region of 0.984 ± 0.Significance. We demonstrate that the proposed method generates sCT images that resemble visual characteristics of a real CT image and has a quantitative accuracy that suits RT dose planning application. We compare the dose calculation from the proposed sCT and the real CT in a radiation therapy treatment planning setup and show that sCT based planning falls within 0.5% target dose error. The method presented here with an initial dose evaluation makes an encouraging precursor to a broader clinical evaluation of sCT based RT planning on different anatomical regions.

Deep Learning Classifies Low- and High-Grade Glioma Patients with High Accuracy, Sensitivity, and Specificity Based on Their Brain White Matter Networks Derived from Diffusion Tensor Imaging.

Classifying low-grade glioma (LGG) patients from high-grade glioma (HGG) is one of the most challenging tasks in planning treatment strategies for brain tumor patients. Previous studies derived several handcrafted features based on the tumor's texture and volume from magnetic resonance images (MRI) to classify LGG and HGG patients. The accuracy of classification was moderate. We aimed to classify LGG from HGG with high accuracy using the brain white matter (WM) network connectivity matrix constructed using diffusion tensor tractography. We obtained diffusion tensor images (DTI) of 44 LGG and 48 HGG patients using routine clinical imaging. Fiber tractography and brain parcellation were performed for each patient to obtain the fractional anisotropy, axial diffusivity, radial diffusivity, and mean diffusivity weighted connectivity matrices. We used a deep convolutional neural network (DNN) for classification and the gradient class activation map (GRAD-CAM) technique to identify the neural connectivity features focused on by the DNN. DNN could classify both LGG and HGG with 98% accuracy. The sensitivity and specificity values were above 0.98. GRAD-CAM analysis revealed a distinct WM network pattern between LGG and HGG patients in the frontal, temporal, and parietal lobes. Our results demonstrate that glioma affects the WM network in LGG and HGG patients differently.

Glioblastoma with primitive neuronal component: An immunohistochemical study and review of literature.

Glioblastoma with primitive neuronal component, a rare neoplasm, is recognized as a distinct histological pattern of glioblastoma. In this study we report the morphological and immunohistochemical features of three cases of glioblastoma with primitive neuronal component diagnosed at the Institute along with a comprehensive literature review. The cases include: (1) 11-year-old girl with right fronto-parietal lesion, (2) 48-year-old male with right parietal lesion, and (3) 36-year-old male with left fronto-parietal lesion. Case 1 had prior history of glioblastoma. All the cases had classic morphology of glioblastoma along with GFAP-negative and synaptophysin-positive primitive neuronal component. The latter was poorly demarcated from the glial component in case 1, while well-defined in the remaining two. All the three cases exhibited diffuse p53 positivity and a higher MIB-1 labelling index in the neuronal component compared to the glial component. One of them (case 3) was IDH1 R132H-mutant with loss of ATRX expression. None were positive for K27M-mutant H3 or G34R-mutant H3.3. Literature review of 50 published cases of glioblastoma with primitive neuronal component was performed. The age of onset ranged from 3 months to 82 years (mean: 50 years) with M:F of 1.5:1. 18.8% of tumors were IDH-mutant, 87.5% were p53 positive and three cases showed H3F3A gene mutations. There was a greater propensity for neuraxial dissemination, noted in 20% of cases. Overall survival of glioblastoma with primitive neuronal component was similar to that of IDH-wildtype glioblastoma (13 months) which was significantly shorter compared to the overall survival of IDH-mutant glioblastoma (33.6 months).

Unusual imaging appearance of a rare cortical ependymoma mimicking angiocentric glioma.

Cortical ependymomas are uncommon lesions. We report a cortical ependymoma displaying an unusual imaging pattern. The lesion demonstrated a T1 hyperintense rim and limited perilesional edema but lacked contrast enhancement. T1 hyperintense rim and stalk-like ventricular extension of FLAIR hyperintensity have previously been considered pathognomonic of angiocentric glioma. Hence, we conclude that the radiologists should be aware of this uncommon imaging pattern of cortical ependymoma. The condition warrants prompt surgical management in view of the increased potential for higher grade transformation, unlike grade I roman numeral may be given angiocentric gliomas.

Machine learning methods for the classification of gliomas: Initial results using features extracted from MR spectroscopy.

CONTEXT: With the advent of new imaging modalities, radiologists are faced with handling increasing volumes of data for diagnosis and treatment planning. The use of automated and intelligent systems is becoming essential in such a scenario. Machine learning, a branch of artificial intelligence, is increasingly being used in medical image analysis applications such as image segmentation, registration and computer-aided diagnosis and detection. Histopathological analysis is currently the gold standard for classification of brain tumors. The use of machine learning algorithms along with extraction of relevant features from magnetic resonance imaging (MRI) holds promise of replacing conventional invasive methods of tumor classification. AIMS: The aim of the study is to classify gliomas into benign and malignant types using MRI data. SETTINGS AND DESIGN: Retrospective data from 28 patients who were diagnosed with glioma were used for the analysis. WHO Grade II (low-grade astrocytoma) was classified as benign while Grade III (anaplastic astrocytoma) and Grade IV (glioblastoma multiforme) were classified as malignant. METHODS AND MATERIALS: Features were extracted from MR spectroscopy. The classification was done using four machine learning algorithms: multilayer perceptrons, support vector machine, random forest and locally weighted learning. RESULTS: Three of the four machine learning algorithms gave an area under ROC curve in excess of 0.80. Random forest gave the best performance in terms of AUC (0.911) while sensitivity was best for locally weighted learning (86.1%). CONCLUSIONS: The performance of different machine learning algorithms in the classification of gliomas is promising. An even better performance may be expected by integrating features extracted from other MR sequences.

Early risk and predictors of cerebrovascular and cardiovascular events in transient ischemic attack and minor ischemic stroke.

BACKGROUND: Transient ischemic attack (TIA) and minor ischemic stroke (MIS) are associated with early recurrence and deterioration respectively. The aim of the present study was to assess the risk of new cerebrovascular and cardiovascular events in a prospective, emergently enrolled patient cohort with TIA and MIS and the predictors of risk. MATERIALS AND METHODS: Patients with TIA and MIS (NIH Stroke Scale [NIHSS] ≤ 5) presenting within the first 48 h between July 2008-June 2009 were prospectively enrolled. The primary outcome was new-onset stroke, TIA, cardiovascular events and vascular death at 90 days and early deterioration in patients with minor stroke. The 90-day outcome was also assessed (excellent outcome; modified Rankin scale [mRS] ≤2). RESULTS: Eighteen (15.3%) of the 118 patients enrolled developed new cerebrovascular or cardiovascular events during the 90 days of follow-up, nine (50%) of which occurred within seven days. Of the all new events 5.9% (7/118) had new stroke, 4.2% (5/118) patients developed early deterioration, 2.5% (3/118) patients had recurrent TIA and 2.5% (3/118) had cardiovascular events at 90 days. Eight (6.7%) patients had poor outcome at 90 days (mRS>2). The factors predicting new vascular events were presence of coronary artery disease (CAD), and stroke etiology being large artery atherosclerosis (LAA). CONCLUSION: In patients with TIA and MIS, despite urgent evaluation and aggressive management, the short-term risk of stroke and other vascular events is high. Those with CAD and LAA should be monitored closely for early deterioration.