Magnetic Resonance Signal Characteristics Research Articles

Fully automatic mpMRI analysis using deep learning predicts peritumoral glioblastoma infiltration and subsequent recurrence.

Glioblastoma (GBM) is most aggressive and common adult brain tumor. The standard treatments typically include maximal surgical resection, followed adjuvant radiotherapy and chemotherapy. However, the efficacy of these treatment is often limited, as tumor often infiltrate into the surrounding brain tissue, often extending beyond the radiologically defined margins. This infiltration contributes to the high recurrence rate and poor prognosis associated with GBM patients, necessitating advanced methods for early and accurate detection of tumor infiltration. Despite the great promise traditional supervised machine learning shows in predicting tumor infiltration beyond resectable margins, these methods are heavily reliant on expert-drawn Regions of Interest (ROIs), which are used to construct multi-variate models of different Magnetic Resonance (MR) signal characteristics associated with tumor infiltration. This process is both time consuming and resource intensive. Addressing this limitation, our study proposes a novel integration of fully automatic methods for generating ROIs with deep learning algorithms to create predictive maps of tumor infiltration. This approach uses pre-operative multi-parametric MRI (mpMRI) scans, encompassing T1, T1Gd, T2, T2-FLAIR, and ADC sequences, to fully leverage the knowledge from previously drawn ROIs. Subsequently, a patch based Convolutional Neural Network (CNN) model is trained on these automatically generated ROIs to predict areas of potential tumor infiltration. The performance of this model was evaluated using a leave-one-out cross-validation approach. Generated predictive maps binarized for comparison against post-recurrence mpMRI scans. The model demonstrates robust predictive capability, evidenced by the average cross-validated accuracy of 0.87, specificity of 0.88, and sensitivity of 0.90. Notably, the odds ratio of 8.62 indicates that regions identified as high-risk on the predictive map were significantly more likely to exhibit tumor recurrence than low-risk regions. The proposed method demonstrates that a fully automatic mpMRI analysis using deep learning can successfully predict tumor infiltration in peritumoral region for GBM patients while bypassing the intensive requirement for expert-drawn ROIs.

Granular cell tumor of soft tissues: MR findings

Granular cell tumors (GrCT) are histologically unique neoplasms that often manifest as painless solitary nodules located in the dermis or submucosal tissue. In this article we analyze the radiologic features of this type of tumor in five patients by magnetic resonance imaging (MRI), analyzing morphologic and signal characteristics. MR imaging findings were similar to other publications and we demonstrate low ADC values in our series.

Tumor de células granulares de partes blandas: hallazgos en resonancia magnética

Granular cell tumors are histologically unique neoplasms that often manifest as painless solitary nodules located in the dermis or submucosal tissue. In this article we analyze the radiologic features of this type of tumor in five patients by magnetic resonance imaging (MRI), analyzing morphologic and signal characteristics. MR imaging findings were similar to other publications and we demonstrate low ADC values in our series.

MRI-based texture analysis to differentiate the most common parotid tumours

To evaluate magnetic resonance imaging (MRI) features and signal characteristics of parotid masses and investigate the added role of texture analysis (TA) in the differentiation of parotid tumours. Ninety-five patients (42 women, 53 men; mean age 51.67±14.15) were included in this study. The study group consisted of 40 pleomorphic adenoma, 45 Warthin's tumour, and 10 mucoepidermoid carcinomas. Two reviewers assessed the MRI sequences retrospectively. Fat-suppressed T2-weighted and contrast-enhanced T1-weighted axial images were used for TA. Receiver operating characteristic curve analyses were performed to evaluate the ability to make a diagnosis. Logistic regression analyses were conducted to explore the independent risk factors among the MRI features and to analyse the added value of TA to the qualitative analysis. Significant differences were found in the tumour border (p<0.001), infiltration of the surrounding tissue (p=0.003), contrast-enhancement grading (p<0.001), perineural spread (p=0.013), and pathological lymph nodes (p<0.001) between the malignant and benign tumours. Kurtosis on contrast-enhanced T1-weighted images, and skewness and kurtosis on T2-weighted images were significantly different between the three groups (p=0.020, <0.001, 0.003; respectively). A kurtosis value on T2-weighted images <2.815 along with an ill-defined border had the highest specificity (98.8%) and positive predictive value (83.3%) in the differentiation of malignant tumours. The addition of TA parameters to the MRI findings may contribute to distinguish benign from malignant parotid tumours.

Magnetic Resonance Signal Characteristics and Experimental Study: Mn(II)-Pectin as an Oral Dual Contrast Agent

This study aimed to investigate the feasibility of using Mn(II)-Pectin as an oral abdominal magnetic resonance contrast agent. The Mn(II)-Pectin solution was prepared and diluted to 11 different concentrations and used for magnetic resonance imaging plain scans. The plain scans were performed with an Mn(II)-Pectin solution (1.594 mmol/L) of the same concentration but at different pH (1–7) values or the same pH (pH = 5.4). The contrast agent may be diluted for further animal studies. The scanned sequences were SET1WI, FRFSE T2WI, PDWI, and Flair T2WI. The signal intensity was measured, and the region of interest was selected as 20 mm2. When the Mn(II)-Pectin concentration ranged from 0.1328 mmol/L to 0.3985 mmol/L, the imaging signal of T1WI was strong and stable (undiluted). When the same manganese ion concentration was used, the Mn(II)-Pectin solution shortened the T1, and the T2 relaxation time was stronger than that for pectin and manganese chloride (MnCl2). The changes in signal intensity were similar at different pH values. With T2WI, the signal intensity did not change with the pH. At the same pH (i.e., pH = 5.4), the signal intensity of Mn(II)-Pectin was significantly higher than for MnCl2 on T1WI, the signal intensity of MnCl2 and Mn(II)-Pectin was almost zero on T2WI, and the signal intensity of pectin was significantly higher than that of MnCl2 and Mn(II)-Pectin on PDWI. The Mn(II)-Pectin solution showed short T1 and T2 signal intensities. Its imaging characteristics were similar to those of the MnCl2 solution. The Mn(II)-Pectin solution is not an ordinary indirect blend but is at least a chemical bond change.

Evolution of Magnetic Resonance Signal Characteristics in Inferior Olivary Nuclei in Patients Presenting with Palatal Myoclonus Syndrome – Cases (P5.6-030)

Evolution of Magnetic Resonance Signal Characteristics in Inferior Olivary Nuclei in Patients Presenting with Palatal Myoclonus Syndrome – Cases (P5.6-030)

Correlation of magnetic resonance signal characteristics and perfusion parameters assessed by volume perfusion computed tomography in hepatocellular carcinoma: Impact on lesion characterization.

To find out if magnetic resonance (MR)-signal characteristics of hepatocellular carcinomas (HCC) correlate with perfusion parameters assessed by volume perfusion computed tomography (VPCT). From October 2009 to January 2014, 26 (mean age, 69.3 years) patients with 36 HCC lesions who underwent both VPCT and MR liver imaging were analysed. We compared signal intensity in the T1w- and T2w-images and wash-in/wash-out kinetics on post-contrast MR images with mean values of blood flow (BF, mL/100 mL per minute), blood volume (BV, mL/100 mL), k-trans (mL/100 mL per minute), arterial liver perfusion (mL/100 mL per minute), portal venous perfusion and hepatic perfusion index (HPI, %) obtained by VPCT. Signal intensity on magnetic resonance imaging (MRI) was classified hyper/iso/hypointense compared with surrounding liver parenchyma. Signal intensity on native T1w- and T2w-images was hyper/iso/hypo in 4/16/16 and 21/14/1 lesions, respectively. Wash-in and wash-out contrast kinetics were found on MRI in 33 of 36 lesions (91.7%) and 25 of 36 lesions (69.4%), respectively. The latter was observed significantly more often in higher graded lesions (P < 0.005). HPI was 94.7% ± 6.5%. There was no significant relationship between lesion's MR-signal intensity, MR signal combinations, size and any of the VPCT-perfusion parameters. However HPI was constantly high in all HCC lesions. VPCT parameters add limited value to MR-lesion characterization. However in HCC lesions with atypical MR signal characteristics HPI can add a parameter to ensure HCC diagnosis.

MR Anatomy of Deep Brain Nuclei with Special Reference to Specific Diseases and Deep Brain Stimulation Localization

Diseases affecting the basal ganglia and deep brain structures vary widely in etiology and include metabolic, infectious, ischemic, and neurodegenerative conditions. Some neurologic diseases, such as Wernicke encephalopathy or pseudohypoparathyroidism, require specific treatments, which if unrecognized could lead to further complications. Other pathologies, such as hypertrophic olivary degeneration, if not properly diagnosed may be mistaken for a primary medullary neoplasm and create unnecessary concern. The deep brain structures are complex and can be difficult to distinguish on routine imaging. It is imperative that radiologists first understand the intrinsic anatomic relationships between the different basal ganglia nuclei and deep brain structures with magnetic resonance (MR) imaging. It is important to understand the "normal" MR signal characteristics, locations, and appearances of these structures. This is essential to recognizing diseases affecting the basal ganglia and deep brain structures, especially since most of these diseases result in symmetrical, and therefore less noticeable, abnormalities. It is also crucial that neurosurgeons correctly identify the deep brain nuclei presurgically for positioning deep brain stimulator leads, the most important being the subthalamic nucleus for Parkinson syndromes and the thalamic ventral intermediate nucleus for essential tremor. Radiologists will be able to better assist clinicians in diagnosis and treatment once they are able to accurately localize specific deep brain structures.

Hippocampal T2 hyperintensities on 7Tesla MRI

Hippocampal focal T2 hyperintensities (HT2Hs), also referred to as hippocampal sulcal cavities, are a common finding on Magnetic Resonance (MR) images. There is uncertainty about their etiology and clinical significance. In this study we aimed to describe these HT2Hs in more detail using high resolution 7 Tesla MR imaging, addressing 1) the MR signal characteristics of HT2Hs, 2) their occurrence frequency, 3) their location within the hippocampus, and 4) their relation with age. We also performed an explorative post-mortem study to examine the histology of HT2Hs.Fifty-eight persons without a history of invalidating neurological or psychiatric disease (mean age 64 ± 8 years; range 43–78 years), recruited through their general practitioners, were included in this study. They all underwent 7 Tesla MRI, including a T1, T2, and FLAIR image. MR signal characteristics of the HT2Hs were assessed on these images by two raters. Also, the location and number of the HT2Hs were assessed. In addition, four formalin-fixed brain slices from two subjects were scanned overnight. HT2Hs identified in these slices were subjected to histopathological analysis.HT2Hs were present in 97% of the subjects (median number per person 10; range 0–20). All HT2Hs detected on the T2 sequence were hypointense on T1 weighted images. Of all HT2Hs, 94% was hypointense and 6% hyperintense on FLAIR. FLAIR hypointense HT2Hs were all located in the vestigial sulcus of the hippocampus, FLAIR hyperintense HT2Hs in the hippocampal sulcus or the gray matter. Post-mortem MRI and histopathological analysis suggested that the hypointense HT2Hs on FLAIR were cavities filled with cerebrospinal fluid. A hyperintense HT2H on FLAIR proved to be a microinfarct upon microscopy.In conclusion, hippocampal T2Hs are extremely common and unrelated to age. They can be divided into two types (hypo- and hyperintense on FLAIR), probably with different etiology.

Glioblastoma Multiforme versus Solitary Supratentorial Brain Metastasis: Differentiation Based on Morphology and Magnetic Resonance Signal Characteristics

To evaluate the diagnostic potential of a multi-factor analysis of morphometric parameters and magnetic resonance (MR) signal characteristics of a mass and peritumoral area to distinguish solitary supratentorial metastasis from glioblastoma multiforme (GBM). MR examinations of 51 patients with histologically proven GBM and 44 with a single supratentorial metastasis were evaluated. A large variety of morphologic criteria and MR signal characteristics in different sequences were analyzed. The data were subjected to logistic regression to investigate their ability to discriminate between GBM and cerebral metastasis. Receiver-operating characteristic (ROC) analysis was used to select an optimal cut-off point for prediction and to assess the predictive value in terms of sensitivity, specificity, and accuracy of the final model. The logistic regression analysis revealed that the ratio of the maximum diameter of the peritumoral area measured on T2-weighted images (d T2) to the maximum diameter of the enhancing mass area (d T1, post-contrast) is the only useful criterion to distinguish single supratentorial brain metastasis from GBM with a lower ratio favoring GBM (accuracy 68 %, sensitivity 84 % and specificity 45 %). The cut-off point for the ratio d T2/d T1 post-contrast was calculated as 2.35. Measurement of maximum diameters of the peritumoral area in relation to the enhancing mass can be evaluated easily in the clinical routine to discriminate GBM from solitary supratentorial metastasis with an accuracy comparable to that of advanced MRI techniques.

Design, construction, and evaluation of a dynamic MR compatible cardiac left ventricle model

Development of magnetic resonance (MR) sequences is important to answer clinical questions and to overcome current limitations. To meet the challenges of cardiac MR, dynamic and reproducible testing conditions are required. We aimed at developing a dynamic MR-compatible cardiac left ventricle model that imitates myocardial tissue properties and simulates dynamic motion. A dynamic left ventricle silicone model was designed to match myocardial T(1) and T(2) relaxation times. Silicone mixtures were explored to replicate T(2) values of myocardial edema. A controllable piston pump was constructed to produce pulsatile flow paradigms. They were validated against flow sensors and MR data, including SSFP-based and phase-contrast-based sequences. A dedicated software interface was developed for the control. Model dimensions represented cardiac left ventricle dimensions of healthy men. The range of end diastolic volumes was 85-175 ml, depending on the driven stroke volume. Stroke volume quantification for flow paradigms of 30∕60∕90∕120 ml resulted in 29.2∕57.6∕88.8∕118.4 ml by MR volumetry, 29.6∕59.9∕89.4∕119.0 ml by phase contrast measurements, and 29.9∕60.4∕91.1∕120.9 ml by flow meter revealing consistency. The system accurately replicated physiological and pathophysiological flow paradigms. The silicon model exhibited T(1) of 1002 ± 8 ms, T(2) of 58 ± 1 ms. Signal intensities (a.u.) of the ventricle model were 128 ± 23 for FGRE (vs 138 ± 17 in vivo) and 1156 ± 37 for b-SSFP (vs 991 ± 96 in vivo). T(2) of 75 ± 2 ms was achieved for the myocardial pathology. We developed a controllable left ventricle model resembling MR signal characteristics of human myocardium, including pathological conditions, and allowing for the replication of contraction and flow paradigms.

Conventional MRI as a diagnostic and prognostic tool in spinal cord injury: a systemic review of its application to date and an overview on emerging MRI methods

The diagnosis and prognosis of traumatic spinal cord injury has historically relied on clinical examination whereby those presenting with severe injuries were deemed unlikely to recover and those presenting with mild injuries were deemed more likely to recover. With the widespread use of MRI to visualize traumatic injury to the spinal cord, a spectrum of previously unseen characteristics ranging from mild T2-weighted signal intensity to complete spinal cord transection is now available to aid in both the diagnosis and prognosis. In this systematic review, the authors outline how clinical examination (using the American Spinal Injury Association standards) and MRI characteristics can be used to classify and characterize acute traumatic cervical spinal cord injury. The reader will gain an appreciation for the different magnetic resonance signal characteristics that can be used to predict a favorable or unfavorable prognosis following traumatic spinal cord injury. The accuracy of this information, in terms of sensitivity and specificity, is presented. Using likelihood ratios, the authors work through specific examples. The use of MRI in the evaluation of the human spinal cord has aided our understanding of the condition significantly. However, there are still several challenges that need to be met, in particular the use of MRI to detect functional abnormalities as well as structural ones. In the coming years, our ability to define damaged circuits in the spinal cord will mean that it will be possible to link structure to function in an objective non-invasive way, which will have implications for the understanding and potential treatment of spinal cord injury.

Tumors and Tumor-Like Lesions of Peripheral Nerves

The diagnosis of a peripheral nerve tumor can often be suggested on imaging. Direct continuity with a neural structure or location along a typical nerve distribution, shape, and intrinsic magnetic resonance (MR) signal characteristics represent the most important signs in this regard. Although several nonneoplastic nerve lesions can be specifically diagnosed by MR imaging, benign and malignant neoplasms of peripheral nerves can usually not be distinguished with confidence. This article reviews the MR imaging appearance, clinical and pathological features of schwannoma, localized neurofibroma, plexiform neurofibroma, intraneural perineurioma, fibrolipomatous hamartoma, nerve sheath ganglion, traumatic neuroma, malignant peripheral nerve sheath tumor, and secondary malignant neoplasms of peripheral nerves. Typical findings are illustrated on the basis of histologically confirmed cases.

MR Signal Characteristics of Viable and Apoptotic Human Mesenchymal Stem Cells in Matrix-Associated Stem Cell Implants for Treatment of Osteoarthritis

To compare magnetic resonance (MR) signal characteristics of contrast agent-labeled apoptotic and viable human mesenchymal stem cells (hMSCs) in matrix-associated stem cell implants. hMSCs were labeled with Food and Drug Administration-approved ferumoxides nanoparticles. One group (A) remained untreated whereas a second group (B) underwent mitomycin C-induced apoptosis induction. Viability of group A and apoptosis of group B was confirmed by caspase-assays and terminal dUTP nick-end labeling (TUNEL) stains. Labeled viable hMSCs, unlabeled viable hMSCs, labeled apoptotic hMSCs, and unlabeled apoptotic hMSCs (n = 7 samples each) in an agarose scaffold were implanted into cartilage defects of porcine patellae specimens and underwent MR imaging at 7 T, using T1-weighted spin-echo sequences, T2-weighted spin-echo sequences, and T2*-weighted gradient-echo sequences. Signal-to-noise ratios (SNR) of the implants were calculated and compared between different experimental groups using linear mixed regression models. Ferumoxides-labeled hMSCs provided a strong negative T2 and T2*-enhancement. Corresponding SNR data of labeled hMSCs were significantly lower compared with unlabeled controls (P < 0.05). Apoptosis induction resulted in a significant signal decline of ferumoxides-labeled hMSC transplants on short echo time T2-weighted spinecho sequences. SNR data of labeled apoptotic hMSCs were significantly lower compared with labeled viable hMSCs (P < 0.05). Apoptosis of transplanted ferumoxides-labeled stem cells in cartilage defects can be visualized noninvasively by a significant signal decline on T2-weighted MR images. The described MR signal characteristics may serve as a noninvasive outcome measure for the assessment of matrix-associated stem cell implants in clinical practice. Additional studies are needed to further enhance the observed differences between viable and apoptotic cells, for example, by further optimizing the applied MR pulse sequence parameters or intracellular contrast agent concentration.

Magnetic Resonance Signal Characteristics of the Cervix as Pregnancy Advances

The objective of this study is to describe magnetic resonance (MR) signal intensity (SI) changes in the cervix during pregnancy. This is an observational cohort study of women with a history of preterm delivery. MR imaging sequences were performed every 3 to 4 weeks. Using 8 regions of interest, the SIs are quantified and analyzed with respect to gestational age. Twenty-seven MR studies were performed on a cohort of 8 women. The SIs of the external os are significantly greater than those of the internal os ( P = .035). Similarly, the SIs of the outer stroma are greater than those of the inner stroma (P = .002). As gestational age advances, the inner to outer stromal SI ratio increases, primarily because of a decreasing SI in the outer stromal layer (P = .03). The MR SIs of the cervical stromal zones display variability during pregnancy and decrease with advancing gestation.

Differential Diagnostic Approach to MR Imaging of White Matter Diseases

The purpose of this article is to explore the magnetic resonance (MR) imaging features and the advanced MR techniques for differential diagnosis of white matter diseases of the brain. Magnetic resonance imaging is exquisitely sensitive for detecting brain abnormalities. Particularly in the evaluation of white matter diseases, MR far outperforms any other imaging technique. Lesions that may be quite subtle or even invisible on computed tomography are often clearly seen on the MR scan. The MR signal characteristics of white matter lesions are similar and relatively nonspecific, but other distinguishing features are often present to assist in diagnosis, such as the pattern of the abnormality, location, signal intensities, and enhancement features. Advanced MR techniques, such as diffusion-weighted imaging, MR spectroscopy, and perfusion imaging, provide additional specificity to the diagnosis of demyelinating diseases. Conventional MR images and advanced MR techniques are very helpful in distinguishing various types of white matter disease, for assessing disease burden, and for separating acute and chronic lesions.

Intramedullary and extramedullary fat globules on magnetic resonance imaging as a diagnostic sign for osteomyelitis

We retrospectively studied the frequency of persistent foci of fat signal on magnetic resonance (MR) imaging in osteomyelitis to assess its frequency, cause and diagnostic value. The radiographs and MR scans of 100 patients with a final diagnosis of osteomyelitis referred to a specialist orthopaedic oncology service with the presumptive diagnosis of a bone tumour were reviewed. The MR signal and morphological characteristics were recorded with particular attention to the presence of persistent fat signal within the infected area, which was classified as diffuse or focal. Seventeen cases were classified on radiographic grounds as acute, 63 as subacute and 20 as chronic osteomyelitis. In the acute group 12 (70%) showed replacement of the marrow with fluid containing residual fatty signal, diffuse in seven and focal in five cases. Two cases showed predominantly fatty marrow with very early marrow oedema and three cases (18%) showed replacement of marrow fat with fluid and no residual fatty foci. None of the subacute group showed foci of fatty signal and two cases of inactive sclerosing osyeomyelitis in the chronic group showed restoration of normal marrow. Persistent fatty signal within the bone as well as soft tissues on MR imaging is a frequent finding in acute osteomyelitis. Radiological-pathological correlation suggests that the increasing intramedullary pressure leads to septic necrosis with death of the lipocytes and release of free fatty globules. This characteristic, but not pathognomonic, MR finding supports the diagnosis of osteomyelitis and may help to exclude the presence of a tumour.

Magnetic resonance signal alterations in the acute onset of heterotopic ossification in patients with spinal cord injury

The purpose of our study was to evaluate magnetic resonance (MR) signal characteristics of acutely forming heterotopic ossification (HO) in paralyzed patients. Fourteen patients with spinal cord injury (female n=2, male n=12, mean age 38.3 years) and acute onset of radiographically proven HO had contrast-enhanced 1.5-T MRI within 13.4+/-18.3 days of clinical onset of symptoms. MR signal alterations of affected muscles, fascia, subcutaneous tissue, skin and adjacent bone were evaluated. A diffuse T2-hyperintense signal of multiple muscle groups was seen in all patients (bilateral in 12) involving quadriceps (n=13, 93%), adductors (n=13, 93%) and iliopsoas (n=12, 86%) with contrast enhancement in n=11 (79%), n=8 (57%) and n=8 (57%) patients. All patients had nonenhancing areas (mean size 2 x 3.5 x 5.8 cm) within diffusely enhancing muscles. HO formation occurred around these nonenhancing areas in four patients with computed tomography follow-up. Other MR findings included fascial edema (n=14, 100%), fascial enhancement (n=13, 93%), subcutaneous edema (n=13, 93%), subcutaneous enhancement (n=12, 86%), bone marrow edema (n=5, 36%), and joint effusion (n=12, 86%). MRI reveals mostly bilateral edema and enhancement of muscles, fascia and subcutaneous tissue during acute onset of HO. HO develops in the periphery of well-defined areas of no enhancement.

MR imaging of avascular scaphoid nonunion before and after vascularized bone grafting

To investigate the magnetic resonance (MR) imaging appearances of chronic nonunion of the scaphoid with proximal pole avascular necrosis before and after insertion of a vascularized bone graft, using computed tomography (CT) as the imaging gold standard. A retrospective study was performed involving MR imaging (n=26), CT scans (n=37) and radiographs (n=52) of 13 men (mean age 29 years, age range 20-38 years) with avascular scaphoid nonunion. Avascular necrosis of the scaphoid proximal pole was confirmed intraoperatively (n=13). MR images were acquired preoperatively and following placement of a vascularized bone graft. Scaphoid MR signal characteristics were assessed for evidence of vascular bone graft incorporation and revascularization of the bone marrow of the proximal pole of the scaphoid and compared with the gold standard of CT. Surgical and clinical notes were reviewed with a minimum 3 year imaging and clinical follow-up in all patients. Graft incorporation with revascularization of the proximal pole of the scaphoid was documented in 9 patients (69%). Graft failure with persistent pseudoarthrosis and avascular necrosis of the scaphoid was seen in 4 patients (31%). MR imaging is useful to determine whether vascularized bone graft incorporation and revascularization of the proximal pole of the scaphoid has occurred in the setting of avascular scaphoid nonunion.

Imaging characteristics of primary cranial Ewing sarcoma

Ewing sarcoma accounts for 10-15% of all childhood malignant bone tumours and is second in prevalence to osteosarcoma. The skull bones are an unusual site of origin of primary Ewing sarcoma in children. Previous reports concentrate on the neurosurgical aspects and relatively good outcome compared to other bone tumours of the calvarium. Reported cases mainly describe the imaging characteristics on CT. To describe the MRI and CT features of primary cranial Ewing sarcoma. The neuroimaging of three cases of primary cranial Ewing sarcoma are reviewed. Our three cases show an extra-axial mass that is high attenuation on CT and low signal on T2-weighted MRI. Haemorrhagic components, dural extension and contrast enhancement are also characteristic features. CT attenuation and magnetic resonance signal characteristics reflect sheets of densely packed cells seen in Ewing sarcoma.