Machine Learning Detects Symptomatic Plaques in Patients With Carotid Atherosclerosis on CT Angiography.

BACKGROUND: This study aimed to develop and validate a computed tomography angiography based machine learning model that uses plaque composition data and degree of carotid stenosis to detect symptomatic carotid plaques in patients with carotid atherosclerosis. METHODS: The machine learning based model was trained using degree of stenosis and the volumes of 13 computed tomography angiography derived intracarotid plaque subcomponents (eg, lipid, intraplaque hemorrhage, calcium) to identify plaques associated with cerebrovascular events. The model was internally validated through repeated 10-fold cross-validation and tested on a dedicated testing cohort according to discrimination and calibration. RESULTS: This retrospective, single-center study evaluated computed tomography angiography scans of 268 patients with both symptomatic and asymptomatic carotid atherosclerosis (163 for the derivation set and 106 for the testing set) performed between March 2013 and October 2019. The area-under-receiver-operating characteristics curve by machine learning on the testing cohort (0.89) was significantly higher than the areas under the curve of traditional logit analysis based on the degree of stenosis (0.51, P<0.001), presence of intraplaque hemorrhage (0.69, P<0.001), and plaque composition (0.78, P<0.001), respectively. Comparable performance was obtained on internal validation. The identified plaque components and associated cutoff values that were significantly associated with a higher likelihood of symptomatic status after adjustment were the ratio of intraplaque hemorrhage to lipid volume (≥50%, 38.5 [10.1-205.1]; odds ratio, 95% CI) and percentage of intraplaque hemorrhage volume (≥10%, 18.5 [5.7-69.4]; odds ratio, 95% CI). CONCLUSIONS: This study presented an interpretable machine learning model that accurately identifies symptomatic carotid plaques using computed tomography angiography derived plaque composition features, aiding clinical decision-making.

Random expert sampling for deep learning segmentation of acute ischemic stroke on non-contrast CT.

BACKGROUND: Outlining acutely infarcted tissue on non-contrast CT is a challenging task for which human inter-reader agreement is limited. We explored two different methods for training a supervised deep learning algorithm: one that used a segmentation defined by majority vote among experts and another that trained randomly on separate individual expert segmentations. METHODS: The data set consisted of 260 non-contrast CT studies in 233 patients with acute ischemic stroke recruited from the multicenter DEFUSE 3 (Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke 3) trial. Additional external validation was performed using 33 patients with matched stroke onset times from the University Hospital Lausanne. A benchmark U-Net was trained on the reference annotations of three experienced neuroradiologists to segment ischemic brain tissue using majority vote and random expert sampling training schemes. The median of volume, overlap, and distance segmentation metrics were determined for agreement in lesion segmentations between (1) three experts, (2) the majority model and each expert, and (3) the random model and each expert. The two sided Wilcoxon signed rank test was used to compare performances (1) to 2) and (1) to (3). We further compared volumes with the 24 hour follow-up diffusion weighted imaging (DWI, final infarct core) and correlations with clinical outcome (modified Rankin Scale (mRS) at 90 days) with the Spearman method. RESULTS: The random model outperformed the inter-expert agreement ((1) to (2)) and the majority model ((1) to (3)) (dice 0.51±0.04 vs 0.36±0.05 (P<0.0001) vs 0.45±0.05 (P<0.0001)). The random model predicted volume correlated with clinical outcome (0.19, P<0.05), whereas the median expert volume and majority model volume did not. There was no significant difference when comparing the volume correlations between random model, median expert volume, and majority model to 24 hour follow-up DWI volume (P>0.05, n=51). CONCLUSION: The random model for ischemic injury delineation on non-contrast CT surpassed the inter-expert agreement ((1) to (2)) and the performance of the majority model ((1) to (3)). We showed that the random model volumetric measures of the model were consistent with 24 hour follow-up DWI.

USE-Evaluator: Performance metrics for medical image segmentation models supervised by uncertain, small or empty reference annotations in neuroimaging.

Performance metrics for medical image segmentation models are used to measure the agreement between the reference annotation and the predicted segmentation. Usually, overlap metrics, such as the Dice, are used as a metric to evaluate the performance of these models in order for results to be comparable. However, there is a mismatch between the distributions of cases and the difficulty level of segmentation tasks in public data sets compared to clinical practice. Common metrics used to assess performance fail to capture the impact of this mismatch, particularly when dealing with datasets in clinical settings that involve challenging segmentation tasks, pathologies with low signal, and reference annotations that are uncertain, small, or empty. Limitations of common metrics may result in ineffective machine learning research in designing and optimizing models. To effectively evaluate the clinical value of such models, it is essential to consider factors such as the uncertainty associated with reference annotations, the ability to accurately measure performance regardless of the size of the reference annotation volume, and the classification of cases where reference annotations are empty. We study how uncertain, small, and empty reference annotations influence the value of metrics on a stroke in-house data set regardless of the model. We examine metrics behavior on the predictions of a standard deep learning framework in order to identify suitable metrics in such a setting. We compare our results to the BRATS 2019 and Spinal Cord public data sets. We show how uncertain, small, or empty reference annotations require a rethinking of the evaluation. The evaluation code was released to encourage further analysis of this topic https://github.com/SophieOstmeier/UncertainSmallEmpty.git.

Non-inferiority of deep learning ischemic stroke segmentation on non-contrast CT within 16-hours compared to expert neuroradiologists.

We determined if a convolutional neural network (CNN) deep learning model can accurately segment acute ischemic changes on non-contrast CT compared to neuroradiologists. Non-contrast CT (NCCT) examinations from 232 acute ischemic stroke patients who were enrolled in the DEFUSE 3 trial were included in this study. Three experienced neuroradiologists independently segmented hypodensity that reflected the ischemic core on each scan. The neuroradiologist with the most experience (expert A) served as the ground truth for deep learning model training. Two additional neuroradiologists' (experts B and C) segmentations were used for data testing. The 232 studies were randomly split into training and test sets. The training set was further randomly divided into 5 folds with training and validation sets. A 3-dimensional CNN architecture was trained and optimized to predict the segmentations of expert A from NCCT. The performance of the model was assessed using a set of volume, overlap, and distance metrics using non-inferiority thresholds of 20%, 3 ml, and 3 mm, respectively. The optimized model trained on expert A was compared to test experts B and C. We used a one-sided Wilcoxon signed-rank test to test for the non-inferiority of the model-expert compared to the inter-expert agreement. The final model performance for the ischemic core segmentation task reached a performance of 0.46 ± 0.09 Surface Dice at Tolerance 5mm and 0.47 ± 0.13 Dice when trained on expert A. Compared to the two test neuroradiologists the model-expert agreement was non-inferior to the inter-expert agreement, [Formula: see text]. The before, CNN accurately delineates the hypodense ischemic core on NCCT in acute ischemic stroke patients with an accuracy comparable to neuroradiologists.

Lessons learned from evolving frameworks in adult glioblastoma.

Glioblastoma (GBM) is the most common and aggressive malignant adult brain tumor. Significant effort has been directed to achieve a molecular subtyping of GBM to impact treatment. The discovery of new unique molecular alterations has resulted in a more effective classification of tumors and has opened the door to subtype-specific therapeutic targets. Morphologically identical GBM may have different genetic, epigenetic, and transcriptomic alterations and therefore different progression trajectories and response to treatments. With a transition to molecularly guided diagnosis, there is now a potential to personalize and successfully manage this tumor type to improve outcomes. The steps to achieve subtype-specific molecular signatures can be extrapolated to other neuroproliferative as well as neurodegenerative disorders.

Imaging of Headache Attributed to Vascular Disorders.

Imaging is essential in the diagnosis of vascular causes of headaches. With advances in technology, there are increasing options of imaging modalities to choose from, each with its own advantages and disadvantages. This article will focus on imaging pearls and pitfalls of vascular causes of headaches. These include aneurysms, vasculitides, vascular malformations, and cerebral venous thrombosis.

Complete Response of a Patient With a Mismatch Repair Deficient Aggressive Pituitary Adenoma to Immune Checkpoint Inhibitor Therapy: A Case Report.

BACKGROUND AND IMPORTANCE: Aggressive pituitary adenomas (APAs) are pituitary tumors that are refractory to standard treatments and carry a poor prognosis. Current treatment guidelines are not standardized but combine surgical resection, radiation therapy, and chemotherapy. Temozolomide is the only chemotherapeutic agent with documented effectiveness and is recommended for APA in European Society of Endocrinology clinical guidelines. CLINICAL PRESENTATION: A 57-year-old man presented with visual deterioration and bitemporal hemianopsia. MRI of the brain demonstrated a sellar mass suspected to be pituitary macroadenoma with displacement of the stalk and optic nerve impingement. The patient underwent stereotactic endoscopic transsphenoidal resection of the mass. Postoperative MRI demonstrated gross total resection. Pathology revealed a sparsely granulated corticotroph adenoma with malignant transformation. Immunohistochemistry showed loss of expression of MLH1 and PMS2 in the tumor cells. Proton therapy was recommended given an elevated Ki67 index and p53 positivity. Before radiotherapy, there was no radiographic evidence of residual tumor. Temozolomide therapy was initiated after surveillance MRI showed recurrence at 16 months postoperatively. However, MRI demonstrated marked progression after 3 cycles. Next-generation sequencing using the MSK-IMPACT platform identified somatic mutations in MLH1 Y548lfs*9 and TP53 R337C . Immunotherapy with ipilimumab/nivolumab was initiated, and MRI demonstrated no residual tumor burden 34 months postoperatively. CONCLUSION: APA is a tumor with frequent recurrence and a short median expected length of survival. Here, we demonstrate the utility of immunotherapy in a single case report of APA, with complete resolution of recurrent APA and improved survival compared with life expectancy.

Pearls & Oy-sters: Pivoting Treatment Regimens of Pediatric Atypical Teratoid Rhabdoid Tumors to Optimize Care in Adult ATRT: A Case Report.

Atypical teratoid rhabdoid tumor (ATRT) is a highly malignant embryonal tumor of the CNS, largely affecting pediatric patients, with exceedingly rare cases in adults at an estimated annual incidence of 1/1,000,000. We report a unique case of ATRT in a 43-year-old female patient who first presented with progressive focal headaches. Imaging revealed a sellar mass with suprasellar extensions, which was partially removed via a transsphenoidal resection. The tumor aggressively recurred just 1 month postoperatively. Her care team pursued a novel treatment plan by using a slightly modified COG ACNS 0332 regimen, which involved radiation, followed by 4 cycles of monthly chemotherapy including vincristine, cyclophosphamide, and cisplatin. Hematopoietic stem cells were collected between radiation and chemotherapy in the event that the patient required stem cell salvage therapy postadjuvant chemotherapy. The MRIs taken at 2 and 4 months postrecurrence indicated a substantial decrease in tumor volume, with corresponding clinical improvements to cranial nerve deficits. Given the scarcity of literature on adult cases of ATRT and the lack of a standard of care for these cases, discussing the efficacy of our patient's treatment plan may aid clinical decision making for adult ATRT cases.

Role of imaging in rare COVID-19 vaccine multiorgan complications.

As of September 18th, 2021, global casualties due to COVID-19 infections approach 200 million, several COVID-19 vaccines have been authorized to prevent COVID-19 infection and help mitigate the spread of the virus. Despite the vast majority having safely received vaccination against SARS-COV-2, the rare complications following COVID-19 vaccination have often been life-threatening or fatal. The mechanisms underlying (multi) organ complications are associated with COVID-19, either through direct viral damage or from host immune response (i.e., cytokine storm). The purpose of this manuscript is to review the role of imaging in identifying and elucidating multiorgan complications following SARS-COV-2 vaccination-making clear that, in any case, they represent a minute fraction of those in the general population who have been vaccinated. The authors are both staunch supporters of COVID-19 vaccination and vaccinated themselves as well.

Automated grading of enlarged perivascular spaces in clinical imaging data of an acute stroke cohort using an interpretable, 3D deep learning framework.

Enlarged perivascular spaces (EPVS), specifically in stroke patients, has been shown to strongly correlate with other measures of small vessel disease and cognitive impairment at 1 year follow-up. Typical grading of EPVS is often challenging and time consuming and is usually based on a subjective visual rating scale. The purpose of the current study was to develop an interpretable, 3D neural network for grading enlarged perivascular spaces (EPVS) severity at the level of the basal ganglia using clinical-grade imaging in a heterogenous acute stroke cohort, in the context of total cerebral small vessel disease (CSVD) burden. T2-weighted images from a retrospective cohort of 262 acute stroke patients, collected in 2015 from 5 regional medical centers, were used for analyses. Patients were given a label of 0 for none-to-mild EPVS (< 10) and 1 for moderate-to-severe EPVS (≥ 10). A three-dimensional residual network of 152 layers (3D-ResNet-152) was created to predict EPVS severity and 3D gradient class activation mapping (3DGradCAM) was used for visual interpretation of results. Our model achieved an accuracy 0.897 and area-under-the-curve of 0.879 on a hold-out test set of 15% of the total cohort (n = 39). 3DGradCAM showed areas of focus that were in physiologically valid locations, including other prevalent areas for EPVS. These maps also suggested that distribution of class activation values is indicative of the confidence in the model's decision. Potential clinical implications of our results include: (1) support for feasibility of automated of EPVS scoring using clinical-grade neuroimaging data, potentially alleviating rater subjectivity and improving confidence of visual rating scales, and (2) demonstration that explainable models are critical for clinical translation.

Association Between CT Angiogram Collaterals and CT Perfusion in Delayed Time Windows for Large Vessel Occlusion Ischemic Strokes.

OBJECTIVES: Recent endovascular trials have established the use of CT perfusion (CTP) in endovascular treatment selection for patients with large vessel occlusions (LVO). However, the relationship between CTP and collateral circulation is unclear in delayed time windows. We explored the relationship between CT Angiogram (CTA) collaterals and CTP parameters in delayed time windows (6-24 hours). MATERIALS AND METHODS: We utilized a single institutional, retrospective stroke registry of consecutive patients between May 2016 and May 2018 with anterior LVO with CTA and CTP imaging within 6-24 hours of stroke onset. We graded baseline collaterals on single phase CTA using modified Tan collateral score (0-3) and dichotomized into good (2-3) and poor (0-1) collaterals. We recorded automated CTP parameters, including estimated ischemic core (cerebral blood flow (CBF)<30%), penumbra (Tmax>6 s), and mismatch ratio. We used Mann-Whitney test and linear regression to assess associations. RESULTS: We included 48 patients with median age of 62 years (IQR= 52-72), median core of 17.5 mL (IQR=0-47), and median penumbra of 117.5 mL (IQR= 62-163.5). Patients with good collaterals had smaller median core (0 mL, IQR=0-12 mL vs. 40.5 mL, IQR=15-60 mL) (p < 0.001), smaller median penumbra (83.5 mL, IQR=43-135 mL vs. 142.5 mL, IQR=77-190 mL) (p = 0.04), larger median mismatch ratio (13.7, IQR=5.7-58.0 vs. 3.1, IQR=2.1-5.0) (p < 0.001), and lower median hypoperfusion intensity ratio (0.23, IQR=0-0.44 vs. 0.52, IQR=0.45-0.63) (p < 0.001) than patients with poor collaterals. CONCLUSIONS: In delayed time window LVO patients, good CTA collaterals are significantly associated with smaller CTP core, smaller penumbra, larger mismatch ratio, and lower hypoperfusion intensity ratio. CTA collateral assessment could be a potential valuable surrogate to perfusion imaging, particularly in stroke centers where CTP is unavailable.

Mixing Enhancement of Non-Newtonian Shear-Thinning Fluid for a Kenics Micromixer.

In this work, a numerical investigation was analyzed to exhibit the mixing behaviors of non-Newtonian shear-thinning fluids in Kenics micromixers. The numerical analysis was performed using the computational fluid dynamic (CFD) tool to solve 3D Navier-Stokes equations with the species transport equations. The efficiency of mixing is estimated by the calculation of the mixing index for different cases of Reynolds number. The geometry of micro Kenics collected with a series of six helical elements twisted 180° and arranged alternately to achieve the higher level of chaotic mixing, inside a pipe with a Y-inlet. Under a wide range of Reynolds numbers between 0.1 to 500 and the carboxymethyl cellulose (CMC) solutions with power-law indices among 1 to 0.49, the micro-Kenics proves high mixing Performances at low and high Reynolds number. Moreover the pressure losses of the shear-thinning fluids for different Reynolds numbers was validated and represented.

Complications in COVID-19 patients: Characteristics of pulmonary embolism.

OBJECTIVE: The purpose of this study is to evaluate chest CT imaging features, clinical characteristics, laboratory values of COVID-19 patients who underwent CTA for suspected pulmonary embolism. We also examined whether clinical, laboratory or radiological characteristics could be associated with a higher rate of PE. MATERIALS AND METHODS: This retrospective study included 84 consecutive patients with laboratory-confirmed SARS-CoV-2 who underwent CTA for suspected PE. The presence and localization of PE as well as the type and extent of pulmonary opacities on chest CT exams were examined and correlated with the information on comorbidities and laboratory values for all patients. RESULTS: Of the 84 patients, pulmonary embolism was discovered in 24 patients. We observed that 87% of PE was found to be in lung parenchyma affected by COVID-19 pneumonia. Compared with no-PE patients, PE patients showed an overall greater lung involvement by consolidation (p = 0.02) and GGO (p < 0.01) and a higher level of D-Dimer (p < 0,01). Moreover, the PE group showed a lower level of saturation (p = 0,01) and required more hospitalization (p < 0,01). CONCLUSION: Our study showed a high incidence of PE in COVID-19 pneumonia. In 87% of patients, PE was found in lung parenchyma affected by COVID-19 pneumonia with a worse CT severity score and a greater number of lung lobar involvement compared with non-PE patients. CT severity, lower level of saturation, and a rise in D-dimer levels could be an indication for a CTPA. ADVANCES IN KNOWLEDGE: Certain findings of non-contrast chest CT could be an indication for a CTPA.

Progressive myelopathy in myelin oligodendrocyte glycoprotein antibody-associated disease: A new mimicker of progressive multiple sclerosis?

Background MOG-IgG-associated disease (MOGAD) in adults typically presents as a monophasic or relapsing optic, spinal, or opticospinal neuroinflammatory syndrome. Current recommendations discourage testing for MOG-IgG in patients with clinical or paraclinical findings more typical of MS, or in patients with a progressive clinical course. However, this approach may impede identification of the full phenotypic spectrum of this recently described disorder. Methods We retrospectively reviewed charts of 39 MOG-IgG-seropositive patients from two Ohio-based neuroimmunology centers to identify unusual disease patterns. Those with a progressive course were included in this case series. Results We describe five cases of progressive myelopathy associated with MOG-IgG. Most patients had features suggestive of MS, including typical MRI and cerebrospinal fluid findings. However, MOG-IgG positive patients with progressive myelopathy showed poor response to MS disease modifying therapy and better response to intravenous immunoglobulins similar to previous reports on MOGAD patients. Conclusion MOG-IgG-seropositive patients may present with progressive myelopathy and may have a clinical and radiologic phenotype suggestive of primary progressive or secondary progressive MS, or progressive solitary sclerosis. MOG-IgG testing should be considered in patients with progressive myelopathy, especially if clinically worsening on MS therapy.

Neurofibromatosis Type 2 (NF2) and the Implications for Vestibular Schwannoma and Meningioma Pathogenesis.

Patients diagnosed with neurofibromatosis type 2 (NF2) are extremely likely to develop meningiomas, in addition to vestibular schwannomas. Meningiomas are a common primary brain tumor; many NF2 patients suffer from multiple meningiomas. In NF2, patients have mutations in the NF2 gene, specifically with loss of function in a tumor-suppressor protein that has a number of synonymous names, including: Merlin, Neurofibromin 2, and schwannomin. Merlin is a 70 kDa protein that has 10 different isoforms. The Hippo Tumor Suppressor pathway is regulated upstream by Merlin. This pathway is critical in regulating cell proliferation and apoptosis, characteristics that are important for tumor progression. Mutations of the NF2 gene are strongly associated with NF2 diagnosis, leading to benign proliferative conditions such as vestibular schwannomas and meningiomas. Unfortunately, even though these tumors are benign, they are associated with significant morbidity and the potential for early mortality. In this review, we aim to encompass meningiomas and vestibular schwannomas as they pertain to NF2 by assessing molecular genetics, common tumor types, and tumor pathogenesis.

Tumor Habitat-derived Radiomic Features at Pretreatment MRI That Are Prognostic for Progression-free Survival in Glioblastoma Are Associated with Key Morphologic Attributes at Histopathologic Examination: A Feasibility Study.

PURPOSE: To identify radiomic features extracted from the tumor habitat on routine MR images that are prognostic for progression-free survival (PFS) and to assess their morphologic basis with corresponding histopathologic attributes in glioblastoma (GBM). MATERIALS AND METHODS: In this retrospective study, 156 pretreatment GBM MR images (gadolinium-enhanced T1-weighted, T2-weighted, and fluid-attenuated inversion recovery [FLAIR] images) were curated. Of these 156 images, 122 were used for training (90 from The Cancer Imaging Archive and 32 from the Cleveland Clinic, acquired between December 1, 2011, and May 1, 2018) and 34 were used for validation. The validation set was obtained from the Ivy Glioblastoma Atlas Project database, for which the percentage extent of 11 histologic attributes was available on corresponding histopathologic specimens of the resected tumor. Following expert annotations of the tumor habitat (necrotic core, enhancing tumor, and FLAIR-hyperintense subcompartments), 1008 radiomic descriptors (eg, Haralick texture features, Laws energy features, co-occurrence of local anisotropic gradient orientations [CoLIAGe]) were extracted from the three MRI sequences. The top radiomic features were obtained from each subcompartment in the training set on the basis of their ability to risk-stratify patients according to PFS. These features were then concatenated to create a radiomics risk score (RRS). The RRS was independently validated on a holdout set. In addition, correlations (P < .05) of RRS features were computed, with the percentage extent of the 11 histopathologic attributes, using Spearman correlation analysis. RESULTS: RRS yielded a concordance index of 0.80 on the validation set and constituted radiomic features, including Laws (capture edges, waves, ripple patterns) and CoLIAGe (capture disease heterogeneity) from enhancing tumor and FLAIR hyperintensity. These radiomic features were correlated with histopathologic attributes associated with disease aggressiveness in GBM, particularly tumor infiltration (P = .0044) and hyperplastic blood vessels (P = .0005). CONCLUSION: Preliminary findings demonstrated significant associations of prognostic radiomic features with disease-specific histologic attributes, with implications for risk-stratifying patients with GBM for personalized treatment decisions. Supplemental material is available for this article. © RSNA, 2020.

Prediction of recurrent glioblastoma after laser interstitial thermal therapy: The role of diffusion imaging.

BACKGROUND: Evaluate the utility of diffusion-weighted imaging (DWI) for the assessment of local recurrence of glioblastoma (GBM) on imaging performed 24 h following MRI-guided laser interstitial thermal therapy (LITT). We hypothesize that microscopic peritumoral infiltration correlates with early subtle variations on DWI images and apparent diffusion coefficient (ADC) maps. METHODS: Of 64 patients with GBM treated with LITT, 39 had MRI scans within 24 h after undergoing LITT. Patterns on DWI images and ADC maps 24 h following LITT were correlated with areas of future GBM recurrence identified through coregistration of subsequent MRI examinations. In the areas of suspected recurrence within the periphery of post-LITT lesions, signal intensity values on ADC maps were recorded and compared with the remaining peritumoral ring. RESULTS: Thirty-nine patients with GBM met the inclusion criteria. For predicting recurrent GBM, areas of decreased DWI signal and increased signal on ADC maps within the expected peritumoral ring of restricted diffusion identified 24 h following LITT showed 86.1% sensitivity, 75.2% specificity, and high correlation (r = 0.53) with future areas of GBM recurrence (P < .01). Areas of future recurrence demonstrated a 37% increase in the ADC value (P < .001), compared with findings in the surrounding treated peritumoral region. A significantly greater area under the receiver operating characteristics curve was determined for ADC values (P < .01). CONCLUSIONS: DWI obtained 24 h following LITT can help predict the location of GBM recurrence months before the development of abnormal enhancement. This may alter future treatment planning, perhaps suggesting areas that may be targeted for additional therapy.

Pyogenic brain abscess, ventriculitis and diffuse meningitis with fatal outcome in an adult: Radiologic-pathologic correlation☆,.

Rupture of brain abscesses with evolution into ventriculitis with meningitis may result in sudden and dramatic worsening of the clinical situation. We present a 57-year-old man with such an event and fatal outcome. Multiple imaging modalities including computed tomography and advanced magnetic resonance imaging were correlated with gross specimen and histologic images. The differential diagnosis of multiple lesions with ring enhancement and prominent perifocal edema includes mainly infectious and neoplastic processes, such as brain abscess, metastasis, and multicentric glioblastoma. Pyogenic ventriculitis is an uncommon manifestation of severe intracranial infection that might be clinically obscure. We discuss the characteristic magnetic resonance findings of brain abscess and its complications, including meningitis and ventriculitis with emphasis on the role of diffusion-weighted and fluid-attenuated inversion recovery imaging.