Imaging findings in Giant Cell Arteritis: Don't Turn A Blind Eye To The Obvious!

Giant cell arteritis (GCA) is the most common primary large vessel systemic vasculitis in the western world. Even though the involvement of scalp and intracranial vessels has received much attention in the neuroradiology literature, GCA, being a systemic vasculitis can involve multiple other larger vessels including aorta and its major head and neck branches. Herein, the authors present a pictorial review of the various cranial, extracranial and orbital manifestations of GCA. An increased awareness of this entity may help with timely and accurate diagnosis, helping expedite therapy and preventing serious complications.ABBREVIATIONS: ACR= American College of Rheumatology, AION= Anterior Ischemic Optic Neuropathy, EULAR= European League Against Rheumatism, GCA= Giant Cell Arteritis, LV-GCA= Large vessel GCA, PMR= Polymyalgia Rheumatica, US= Ultrasound, VWI= Vessel Wall Imaging.

Model-based iterative reconstruction for direct imaging with point spread function encoded echo planar MRI.

BACKGROUND: Echo planar imaging (EPI) is a fast measurement technique commonly used in magnetic resonance imaging (MRI), but is highly sensitive to measurement non-idealities in reconstruction. Point spread function (PSF)-encoded EPI is a multi-shot strategy which alleviates distortion, but acquisition of encodings suitable for direct distortion-free imaging prolongs scan time. In this work, a model-based iterative reconstruction (MBIR) framework is introduced for direct imaging with PSF-EPI to improve image quality and acceleration potential. METHODS: An MBIR platform was developed for accelerated PSF-EPI. The reconstruction utilizes a subspace representation, is regularized to promote local low-rankedness (LLR), and uses variable splitting for efficient iteration. Comparisons were made against standard reconstructions from prospectively accelerated PSF-EPI data and with retrospective subsampling. Exploring aggressive partial Fourier acceleration of the PSF-encoding dimension, additional comparisons were made against an extension of Homodyne to direct PSF-EPI in numerical experiments. A neuroradiologists' assessment was completed comparing images reconstructed with MBIR from retrospectively truncated data directly against images obtained with standard reconstructions from non-truncated datasets. RESULTS: Image quality results were consistently superior for MBIR relative to standard and Homodyne reconstructions. As the MBIR signal model and reconstruction allow for arbitrary sampling of the PSF space, random sampling of the PSF-encoding dimension was also demonstrated, with quantitative assessments indicating best performance achieved through nonuniform PSF sampling combined with partial Fourier. With retrospective subsampling, MBIR reconstructs high-quality images from sub-minute scan datasets. MBIR was shown to be superior in a neuroradiologists' assessment with respect to three of five performance criteria, with equivalence for the remaining two. CONCLUSIONS: A novel image reconstruction framework is introduced for direct imaging with PSF-EPI, enabling arbitrary PSF space sampling and reconstruction of diagnostic-quality images from highly accelerated PSF-encoded EPI data.

Dorsal arachnoid web adjacent to paraspinal retained bullet: Blunt post-traumatic etiology.

Dorsal arachnoid webs are uncommon, and of uncertain etiology. We present a case in which imaging findings of a dorsal arachnoid web were identified at the level of a known prior gunshot injury where a retained bullet was lodged adjacent to the spine, without associated penetrating injury to the spine, suggesting blunt post-traumatic etiology.

High-resolution computed tomography angiography of the orbit using a photon-counting computed tomography scanner.

BACKGROUND AND PURPOSE: Recent introduction of photon counting detector (PCD) computed tomography (CT) scanners into clinical practice further improve CT angiography (CTA) depiction of orbital arterial vasculature compared to conventional energy integrating detector (EID) CT scanners. PCD-CTA of the orbit can provide a detailed arterial roadmap of the orbit which can de diagnostic on its own or serve as a helpful planning adjunct for both diagnostic and therapeutic catheter-based angiography of the orbit. METHODS: For this review, EID and PCD-CT imaging was obtained in 28 volunteers. The volume CT dose index was closely matched. A dual-energy scanning protocol was used on EID-CT. An ultra-high-resolution (UHR) scan mode was used on PCD-CT. Images were reconstructed at 0.6 mm slice thickness using a closely matched medium-sharp standard resolution (SR) kernel. High-resolution (HR) images with the sharpest quantitative kernel were also reconstructed on PCD-CT at the thinnest slice thickness of 0.2 mm. A denoising algorithm was applied to the HR image series. RESULTS: The imaging description of the orbital vascular anatomy presented in this work was derived from these patients' PCD-CTA images in combination with review of the literature. We found that orbital arterial anatomy is much better depicted with PCD-CTA, and this work can serve primarily as an imaging atlas of the normal orbital vascular anatomy. CONCLUSION: With recent advances in technology, arterial anatomy of the orbit is much better depicted with PCD-CTA as opposed to EID-CTA. Current orbital PCD-CTA technology approaches the necessary resolution threshold for reliable evaluation of central retinal artery occlusion.

Dorsoventral splitting of the infundibulum in a child with pituitary hypoplasia.

Pituitary development arises from ectodermal tissue creating Rathke's pouch and ultimately the adenohypophysis anteriorly whereas neuroectodermal tissue arising from the diencephalon creates the neurohypophysis posteriorly. Alterations in pituitary development can lead to hormonal dysregulation and dysfunction. Following clinical suspicion of pituitary endocrinopathy, MRI plays a vital role in identifying and characterizing underlying structural abnormalities of the pituitary gland, as well as any associated extrapituitary findings. Here we report a case of an 18-month-old female presenting with short stature and growth hormone deficiency. MRI was notable for a shallow sella turcica, a hypoplastic adenohypophysis, thin pituitary stalk, and ectopic neurohypophysis. Interestingly, the pituitary stalk was noted to split dorsoventrally with a split pituitary bright spot and T1 hypointense lobe hypothesized to represent separation of the posterior pituitary lobes.

Technical note: Improved differentiation of calcification from hemosiderin using paramagnetic- and diamagnetic-specific magnetic resonance susceptibility weighted imaging (p-SWI, d-SWI).

INTRODUCTION: Differentiation of calcification and calcium-containing tissue from blood products remains challenging using magnetic resonance imaging (MRI). We developed a novel post-processing algorithm which creates both paramagnetic- and diamagnetic-specific SWI images generated from T2* weighted images using distinct "positive" and "negative" phase masks. METHODS: 10 patients who had undergone clinical MRI scanning of the brain with a rapid echo planar based T2*-weighted EPI-GRE pulse sequence with evidence for either hemosiderin and/or calcifications were retrospectively identified. Complex raw k-space data from individual imaging coils were then extracted, reconstructed, and appropriately combined to produce magnitude and phase images using a phase preserving method. The final reconstructed images included the T2* EPI-GRE magnitude images, p-SWI and d-SWI images. Filtered phase images were also available for review. Correlation with CT scans and MR imaging appearance over time corroborated the composition of the voxels. RESULTS: Differential "blooming" of diamagnetic and paramagnetic foci was readily identified on the corresponding p-SWI and d-SWI images and provided fast and reliable visual differentiation of diamagnetic from paramagnetic susceptibility effects by ascertaining which of the two images depicted the greatest "blooming" effect. Correlation with the available filtered phase maps was not necessary for differentiation of paramagnetic from diamagnetic image components. CONCLUSION: Clinical interpretation of SWI images can be further enhanced by creating specific p-SWI and d-SWI image pairs which contain greater visual information than the combination of standard p-SWI images and phase image.

Enhanced clinical task-based fMRI metrics through locally low-rank denoising of complex-valued data.

OBJECTIVE: This study investigates a locally low-rank (LLR) denoising algorithm applied to source images from a clinical task-based functional MRI (fMRI) exam before post-processing for improving statistical confidence of task-based activation maps. METHODS: Task-based motor and language fMRI was obtained in eleven healthy volunteers under an IRB approved protocol. LLR denoising was then applied to raw complex-valued image data before fMRI processing. Activation maps generated from conventional non-denoised (control) data were compared with maps derived from LLR-denoised image data. Four board-certified neuroradiologists completed consensus assessment of activation maps; region-specific and aggregate motor and language consensus thresholds were then compared with nonparametric statistical tests. Additional evaluation included retrospective truncation of exam data without and with LLR denoising; a ROI-based analysis tracked t-statistics and temporal SNR (tSNR) as scan durations decreased. A test-retest assessment was performed; retest data were matched with initial test data and compared for one subject. RESULTS: fMRI activation maps generated from LLR-denoised data predominantly exhibited statistically significant (p = 4.88×10-4 to p = 0.042; one p = 0.062) increases in consensus t-statistic thresholds for motor and language activation maps. Following data truncation, LLR data showed task-specific increases in t-statistics and tSNR respectively exceeding 20 and 50% compared to control. LLR denoising enabled truncation of exam durations while preserving cluster volumes at fixed thresholds. Test-retest showed variable activation with LLR data thresholded higher in matching initial test data. CONCLUSION: LLR denoising affords robust increases in t-statistics on fMRI activation maps compared to routine processing, and offers potential for reduced scan duration while preserving map quality.

Evaluation of a New, Highly Flexible Radiofrequency Coil for MR Simulation of Patients Undergoing External Beam Radiation Therapy.

PURPOSE: To evaluate the performance of a new, highly flexible radiofrequency (RF) coil system for imaging patients undergoing MR simulation. METHODS: Volumetric phantom and in vivo images were acquired with a commercially available and prototype RF coil set. Phantom evaluation was performed using a silicone-filled humanoid phantom of the head and shoulders. In vivo assessment was performed in five healthy and six patient subjects. Phantom data included T1-weighted volumetric imaging, while in vivo acquisitions included both T1- and T2-weighted volumetric imaging. Signal to noise ratio (SNR) and uniformity metrics were calculated in the phantom data, while SNR values were calculated in vivo. Statistical significance was tested by means of a non-parametric analysis of variance test. RESULTS: At a threshold of p = 0.05, differences in measured SNR distributions within the entire phantom volume were statistically different in two of the three paired coil set comparisons. Differences in per slice average SNR between the two coil sets were all statistically significant, as well as differences in per slice image uniformity. For patients, SNRs within the entire imaging volume were statistically significantly different in four of the nine comparisons and seven of the nine comparisons performed on the per slice average SNR values. For healthy subjects, SNRs within the entire imaging volume were statistically significantly different in seven of the nine comparisons and eight of the nine comparisons when per slice average SNR was tested. CONCLUSIONS: Phantom and in vivo results demonstrate that image quality obtained from the novel flexible RF coil set was similar or improved over the conventional coil system. The results also demonstrate that image quality is impacted by the specific coil configurations used for imaging and should be matched appropriately to the anatomic site imaged to ensure optimal and reproducible image quality.

Arterial Spin Labeling Applications in Pediatric and Adult Neurologic Disorders.

Arterial spin labeling (ASL) is a powerful noncontrast magnetic resonance imaging (MRI) technique that enables quantitative evaluation of brain perfusion. To optimize the clinical and research utilization of ASL, radiologists and physicists must understand the technical considerations and age-related variations in normal and disease states. We discuss advanced applications of ASL across the lifespan, with example cases from children and adults covering a wide variety of pathologies. Through literature review and illustrated clinical cases, we highlight the subtleties as well as pitfalls of ASL interpretation. First, we review basic physical principles, techniques, and artifacts. This is followed by a discussion of normal perfusion variants based on age and physiology. The three major categories of perfusion abnormalities-hypoperfusion, hyperperfusion, and mixed patterns-are covered with an emphasis on clinical interpretation and relationship to the disease process. Major etiologies of hypoperfusion include large artery, small artery, and venous disease; other vascular conditions; global hypoxic-ischemic injury; and neurodegeneration. Hyperperfusion is characteristic of vascular malformations and tumors. Mixed perfusion patterns can be seen with epilepsy, migraine, trauma, infection/inflammation, and toxic-metabolic encephalopathy. LEVEL OF EVIDENCE: 4 TECHNICAL EFFICACY STAGE: 3.

Improved Brain MR Imaging from a Compact, Lightweight 3T Scanner with High-Performance Gradients.

BACKGROUND: A low-cryogen, compact 3T (C3T) MRI scanner with high-performance gradients capable of simultaneously achieving 80 mT/m gradient amplitude and 700 T/m/second slew rate has been in use to study research patients since March 2016 but has not been implemented in the clinical practice. PURPOSE: To compare head MRI examinations obtained with the C3T system and a conventional whole-body 3T (WB3T) scanner in seven parameters across five commonly used brain imaging sequences. STUDY TYPE: Prospective. SUBJECTS: Thirty patients with a clinically indicated head MRI. SEQUENCE: 3T; T1 FLAIR, T1 MP-RAGE, 3D T2 FLAIR, T2 FSE, and DWI. ASSESSMENT: All patients tolerated the scans well. Three board-certified neuroradiologists scored the comparative quality of C3T and WB3T images in blinded fashion using a five-point Likert scale in terms of: signal-to-noise ratio, lesion conspicuity, motion artifact, gray/white matter contrast, cerebellar folia, susceptibility artifact, and overall quality. STATISTICAL TEST: Left-sided, right-sided, and two-sided Wilcoxon signed rank test; Fisher's method. A P value <0.05 was considered statistically significant. RESULTS: The C3T system performed better than the WB3T in virtually all comparisons, except for motion artifacts for the T1 FLAIR and T1 MP-RAGE sequences, where the WB3T system was deemed better. When combining all sequences together, the C3T system outperformed the WB3T system in all image quality parameters evaluated, except for motion artifact (P = 0.13). DATA CONCLUSION: The C3T scanner provided better overall image quality for all sequences, and performed better in all individual categories, except for motion artifact on the T1 FLAIR and T1 MP-RAGE. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY STAGE: 1.

Implementation and Rationale for a Unified Clinical and Imaging Protocol for Evaluation and Treatment of Moyamoya Angiopathy: A Single Institutional Experience.

Moyamoya disease (MMD) is a complex and incompletely-understood cerebrovascular pathological entity that requires thorough clinical and imaging evaluation. Moyamoya is rare, thereby making the establishment of an effective, thorough and interdisciplinary patient evaluation protocol challenging, even within specialized referral centers. Nevertheless, implementation of such a protocol is crucial in order to provide the best possible evaluation and treatment for MMD patients. Here, we describe our institution's implementation of, rationale for, and experience with a comprehensive multidisciplinary collaboration and evaluation strategy for adult patients with moyamoya. This evaluation course consists of, first of all, a thorough clinical and laboratory evaluation with a vascular neurologist. This is followed by a comprehensive imaging assessment which evaluates angiographic and parenchymal features, in addition to cerebrovascular functionality. Finally, appropriate referrals are made to consulting services as indicated, which includes vascular neurosurgery. These steps are described in detail herein.

Whole-brain 3D MR fingerprinting brain imaging: clinical validation and feasibility to patients with meningioma.

PURPOSE: MR fingerprinting (MRF) is a MR technique that allows assessment of tissue relaxation times. The purpose of this study is to evaluate the clinical application of this technique in patients with meningioma. MATERIALS AND METHODS: A whole-brain 3D isotropic 1mm3 acquisition under a 3.0T field strength was used to obtain MRF T1 and T2-based relaxometry values in 4:38 s. The accuracy of values was quantified by scanning a quantitative MR relaxometry phantom. In vivo evaluation was performed by applying the sequence to 20 subjects with 25 meningiomas. Regions of interest included the meningioma, caudate head, centrum semiovale, contralateral white matter and thalamus. For both phantom and subjects, mean values of both T1 and T2 estimates were obtained. Statistical significance of differences in mean values between the meningioma and other brain structures was tested using a Friedman's ANOVA test. RESULTS: MR fingerprinting phantom data demonstrated a linear relationship between measured and reference relaxometry estimates for both T1 (r2 = 0.99) and T2 (r2 = 0.97). MRF T1 relaxation times were longer in meningioma (mean ± SD 1429 ± 202 ms) compared to thalamus (mean ± SD 1054 ± 58 ms; p = 0.004), centrum semiovale (mean ± SD 825 ± 42 ms; p < 0.001) and contralateral white matter (mean ± SD 799 ± 40 ms; p < 0.001). MRF T2 relaxation times were longer for meningioma (mean ± SD 69 ± 27 ms) as compared to thalamus (mean ± SD 27 ± 3 ms; p < 0.001), caudate head (mean ± SD 39 ± 5 ms; p < 0.001) and contralateral white matter (mean ± SD 35 ± 4 ms; p < 0.001) CONCLUSIONS: Phantom measurements indicate that the proposed 3D-MRF sequence relaxometry estimations are valid and reproducible. For in vivo, entire brain coverage was obtained in clinically feasible time and allows quantitative assessment of meningioma in clinical practice.

Whole brain 3D MR fingerprinting in multiple sclerosis: a pilot study.

BACKGROUND: MR fingerprinting (MRF) is a novel imaging method proposed for the diagnosis of Multiple Sclerosis (MS). This study aims to determine if MR Fingerprinting (MRF) relaxometry can differentiate frontal normal appearing white matter (F-NAWM) and splenium in patients diagnosed with MS as compared to controls and to characterize the relaxometry of demyelinating plaques relative to the time of diagnosis. METHODS: Three-dimensional (3D) MRF data were acquired on a 3.0T MRI system resulting in isotropic voxels (1 × 1 × 1 mm3) and a total acquisition time of 4 min 38 s. Data were collected on 18 subjects paired with 18 controls. Regions of interest were drawn over MRF-derived T1 relaxometry maps encompassing selected MS lesions, F-NAWM and splenium. T1 and T2 relaxometry features from those segmented areas were used to classify MS lesions from F-NAWM and splenium with T-distributed stochastic neighbor embedding algorithms. Partial least squares discriminant analysis was performed to discriminate NAWM and Splenium in MS compared with controls. RESULTS: Mean out-of-fold machine learning prediction accuracy for discriminant results between MS patients and controls for F-NAWM was 65 % (p = 0.21) and approached 90 % (p < 0.01) for the splenium. There was significant positive correlation between time since diagnosis and MS lesions mean T2 (p = 0.015), minimum T1 (p = 0.03) and negative correlation with splenium uniformity (p = 0.04). Perfect discrimination (AUC = 1) was achieved between selected features from MS lesions and F-NAWM. CONCLUSIONS: 3D-MRF has the ability to differentiate between MS and controls based on relaxometry properties from the F-NAWM and splenium. Whole brain coverage allows the assessment of quantitative properties within lesions that provide chronological assessment of the time from MS diagnosis.

Application of Adaptive Image Receive Coil Technology for Whole-Brain Imaging.

OBJECTIVE. The Adaptive Image Receive (AIR) radiofrequency coil is an emergent technology that is lightweight and flexible and exhibits electrical characteristics that overcome many of the limitations of traditional rigid coil designs. The purpose of this study was to apply the AIR coil for whole-brain imaging and compare the performance of a prototype AIR coil array with the performance of conventional head coils. SUBJECTS AND METHODS. A phantom and 15 healthy adult participants were imaged. A prototype 16-channel head AIR coil was compared with conventional 8-and 32-channel head coils using clinically available MRI sequences. During consensus review, two board-certified neuroradiologists graded the AIR coil compared with an 8-channel coil and a 32-channel coil on a 5-point ordinal scale in multiple categories. One- and two-sided Wilcoxon signed rank tests were performed. Noise covariance matrices and geometry factor (g-factor) maps were calculated. RESULTS. The signal-to-noise ratio, structural sharpness, and overall image quality scores of the prototype 16-channel AIR coil were better than those of the 8-channel coil but were not as good as those of the 32-channel coil. Noise covariance matrices showed stable performance of the AIR coil across participants. The median g-factors for the 16-channel AIR coil were, overall, less than those of the 8-channel coil but were greater than those of the 32-channel coil. CONCLUSION. On average, the prototype 16-channel head AIR coil outperformed a conventional 8-channel head coil but did not perform as well as a conventional 32-channel head coil. This study shows the feasibility of the novel AIR coil technology for imaging the brain and provides insight for future coil design improvements.

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.

Photon-counting Detector CT: System Design and Clinical Applications of an Emerging Technology.

Photon-counting detector (PCD) CT is an emerging technology that has shown tremendous progress in the last decade. Various types of PCD CT systems have been developed to investigate the benefits of this technology, which include reduced electronic noise, increased contrast-to-noise ratio with iodinated contrast material and radiation dose efficiency, reduced beam-hardening and metal artifacts, extremely high spatial resolution (33 line pairs per centimeter), simultaneous multienergy data acquisition, and the ability to image with and differentiate among multiple CT contrast agents. PCD technology is described and compared with conventional CT detector technology. With the use of a whole-body research PCD CT system as an example, PCD technology and its use for in vivo high-spatial-resolution multienergy CT imaging is discussed. The potential clinical applications, diagnostic benefits, and challenges associated with this technology are then discussed, and examples with phantom, animal, and patient studies are provided. ©RSNA, 2019.

De novo formation of a carotid web: case report.

A carotid web is a shelf-like intraluminal filling defect typically arising from the posterolateral wall of the proximal internal carotid artery. It is recognized as a possible cause of ischemic stroke in young adults. However, its etiopathogenesis is controversial and remains to be fully elucidated. The authors report de novo formation of a carotid web from an intimal dissection documented on serial imaging studies. The findings in this case suggest that a focal intimal dissection could be the underlying cause of a carotid web. Lower shear stress at the posterolateral wall of the proximal internal carotid artery is hypothesized to be a predisposing factor and explains the predilection of a carotid web for this specific location.

Deep learning algorithms for detection of critical findings in head CT scans: a retrospective study.

BACKGROUND: Non-contrast head CT scan is the current standard for initial imaging of patients with head trauma or stroke symptoms. We aimed to develop and validate a set of deep learning algorithms for automated detection of the following key findings from these scans: intracranial haemorrhage and its types (ie, intraparenchymal, intraventricular, subdural, extradural, and subarachnoid); calvarial fractures; midline shift; and mass effect. METHODS: We retrospectively collected a dataset containing 313 318 head CT scans together with their clinical reports from around 20 centres in India between Jan 1, 2011, and June 1, 2017. A randomly selected part of this dataset (Qure25k dataset) was used for validation and the rest was used to develop algorithms. An additional validation dataset (CQ500 dataset) was collected in two batches from centres that were different from those used for the development and Qure25k datasets. We excluded postoperative scans and scans of patients younger than 7 years. The original clinical radiology report and consensus of three independent radiologists were considered as gold standard for the Qure25k and CQ500 datasets, respectively. Areas under the receiver operating characteristic curves (AUCs) were primarily used to assess the algorithms. FINDINGS: The Qure25k dataset contained 21 095 scans (mean age 43 years; 9030 [43%] female patients), and the CQ500 dataset consisted of 214 scans in the first batch (mean age 43 years; 94 [44%] female patients) and 277 scans in the second batch (mean age 52 years; 84 [30%] female patients). On the Qure25k dataset, the algorithms achieved an AUC of 0·92 (95% CI 0·91-0·93) for detecting intracranial haemorrhage (0·90 [0·89-0·91] for intraparenchymal, 0·96 [0·94-0·97] for intraventricular, 0·92 [0·90-0·93] for subdural, 0·93 [0·91-0·95] for extradural, and 0·90 [0·89-0·92] for subarachnoid). On the CQ500 dataset, AUC was 0·94 (0·92-0·97) for intracranial haemorrhage (0·95 [0·93-0·98], 0·93 [0·87-1·00], 0·95 [0·91-0·99], 0·97 [0·91-1·00], and 0·96 [0·92-0·99], respectively). AUCs on the Qure25k dataset were 0·92 (0·91-0·94) for calvarial fractures, 0·93 (0·91-0·94) for midline shift, and 0·86 (0·85-0·87) for mass effect, while AUCs on the CQ500 dataset were 0·96 (0·92-1·00), 0·97 (0·94-1·00), and 0·92 (0·89-0·95), respectively. INTERPRETATION: Our results show that deep learning algorithms can accurately identify head CT scan abnormalities requiring urgent attention, opening up the possibility to use these algorithms to automate the triage process. FUNDING: Qure.ai.