AI tools in Emergency Radiology reading room: a new era of Radiology.

Artificial intelligence tools in radiology practices have surged, with modules developed to target specific findings becoming increasingly prevalent and proving valuable in the daily emergency room radiology practice. The number of US Food and Drug Administration-cleared radiology-related algorithms has soared from just 10 in early 2017 to over 200 presently. This review will concentrate on the present utilization of AI tools in clinical ER radiology setting, including a brief discussion of the limitations of the technique. As radiologists, it is essential that we embrace this technology, comprehend its constraints, and use it to improve patient care.

Nontraumatic brain parenchymal hemorrhage: The usual suspects and more.

Brain parenchymal hemorrhage is a common neuroimaging finding in an emergency room. It is considered primary in the absence of an underlying lesion or coagulopathy. Secondary hemorrhages are caused by various structural causes and pathologies. The goals of imaging are to identify the hematoma, assess factors that have prognostic significance, assess associated complications, identify an underlying etiology whenever possible, and guide therapeutic decisions. The review provides an illustrative review of various etiologies of non-traumatic brain parenchymal hemorrhage and their imaging evaluation.

Paramagnetic Rim Lesions in Multiple Sclerosis: Comparison of Visualization at 1.5-T and 3-T MRI.

BACKGROUND. Multiple sclerosis (MS) is characterized by both acute and chronic intrathecal inflammation. A subset of MS lesions show paramagnetic rims on susceptibility-weighted MRI sequences, reflecting iron accumulation in microglia. These para-magnetic rim lesions have been proposed as a marker of compartmentalized smoldering disease. Paramagnetic rim lesions have been shown at 7 T and, more recently, at 3 T. As susceptibility effects are weaker at lower field strength, it remains unclear if paramagnetic rim lesions are visible at 1.5 T. OBJECTIVE. The purpose of our study was to compare visualization of paramagnetic rim lesions using susceptibility-weighted imaging at 1.5-T and 3-T MRI in patients with MS. METHODS. This retrospective study included nine patients (five women, four men; mean age, 46.8 years) with MS who underwent both 1.5-T and 3-T MRI using a comparable susceptibility-weighted angiography (SWAN) sequence from the same manufacturer. Lesions measuring greater than 3 mm were annotated. Two reviewers independently assessed images at each field strength in separate sessions and classified the annotated lesions as isointense, diffusely paramagnetic, or paramagnetic rim lesions. Discrepancies were discussed at consensus sessions including a third reviewer. Agreement was assessed using kappa coefficients. RESULTS. Based on the 3-T consensus readings, 115 of 140 annotated lesions (82%) were isointense lesions, 16 (11%) were diffusely paramagnetic lesions, and nine (6%) were paramagnetic rim lesions; based on the 1.5-T consensus readings, 115 (82%) were isointense lesions, 14 (10%) were diffusely paramagnetic lesions, and 11 (8%) were para-magnetic rim lesions. The mean lesion diameter was 11.9 mm for paramagnetic rim lesions versus 6.4 mm for diffusely paramagnetic lesions (p = .006) and 7.8 mm for iso-intense lesions (p = .003). Interrater agreement for lesion classification as a paramagnetic rim lesion was substantial at 1.5 T (κ = 0.65) and 3 T (κ = 0.70). Agreement for paramagnetic rim lesions was also substantial between the consensus readings at the two field strengths (κ = 0.79). CONCLUSION. We show comparable identification of paramagnetic rim lesions at 1.5-T and 3-T MRI with substantial interrater agreement at both field strengths and substantial consensus agreement between the field strengths. CLINICAL IMPACT. Paramagnetic rim lesions may be an emerging marker of chronic neuroinflammation in MS. Their visibility at 1.5 T supports the translational potential of paramagnetic rim lesion identification to more widespread clinical settings, where 1.5-T scanners are prevalent.

Computed tomography angiography-derived area stenosis calculations overestimate degree of carotid stenosis compared with North American Symptomatic Carotid Endarterectomy Trial-derived diameter stenosis calculations.

OBJECTIVE: The degree of carotid artery stenosis, calculated using catheter-based angiography and the North American Symptomatic Carotid Endarterectomy Trial (NASCET) method, has been shown to predict the stroke risk in several, large, randomized controlled trials. In the present era, patients have been increasingly evaluated using computed tomography (CT) angiography (CTA) before carotid artery revascularization, especially as the use of transcarotid artery revascularization has increased. Interpretation of CTA findings regarding the degree of carotid stenosis has not been standardized, with both NASCET methods and the area stenosis used. We performed a single-institution, blinded, retrospective analysis of CTA studies using both the NASCET method and the CT-derived area stenosis to assess the concordance and discordance between the two methods when evaluating ≥70% and ≥80% stenosis. METHODS: The UMass Memorial Medical Center vascular laboratory database was queried for all carotid duplex ultrasound scans performed from 2008 to 2017. The included patients were limited to those with duplex-defined ≥70% stenosis (defined as a peak systolic velocity of ≥125 cm/s and an internal carotid artery/common carotid artery ratio of ≥4), and a correlative CTA study performed within 1 year of the duplex ultrasound examination. A blinded review of all correlative CTA studies using centerline measurements on a three-dimensional workstation (Aquarius iNtuition Viewer; Terarecon, Durham, NC) was performed to characterize the degree of carotid stenosis using the NASCET method and the area stenosis method. Patients were excluded if revascularization had been performed between the two imaging studies. RESULTS: Of the 37,204 carotid duplex ultrasound scans reviewed (performed from 2008 to 2017), 3480 arteries met the criteria for duplex ultrasound-defined ≥70% stenosis. A correlative CTA study within 1 year of the duplex ultrasound examination was identified in 460 arteries, of which 320 were adequate quality for blinded review. The median interval between the duplex ultrasound and CTA examinations was 9.5 days. Concordance between the area stenosis and NASCET methods was poor for both ≥70% (κ = 0.32) and ≥80% (κ = 0.25) stenosis. Of the 247 arteries considered to have ≥70% area stenosis, 127 (51.4%) were considered to have ≥70% stenosis using the NASCET method. Of the 169 arteries considered to have ≥80% area stenosis, 44 (26.0%) were considered to have ≥80% stenosis using the NASCET method. CONCLUSIONS: The area stenosis CTA calculations of carotid artery stenosis dramatically overestimated the degree of carotid stenosis compared with that calculated using the NASCET method. Given that stroke risk estimates have been determined from trials that used the NASCET method, the area stenosis method likely overestimates the risk of stroke. Therefore, area stenosis calculations could lead to unnecessary carotid revascularization procedures. This model highlights the need for standardized usage of the NASCET method when using CTA as the imaging modality to determine the threshold for carotid revascularization.

MRI Findings of Injury to the Longus Colli Muscle in Patients With Neck Trauma.

OBJECTIVE: The longus colli muscle (LCM) forms the bulk of the deep flexor muscles of the neck. To our knowledge, very little information on the effects of trauma on this muscle group has been published. We describe MRI findings related to injury of the LCM in patients with a history of neck trauma. MATERIALS AND METHODS: A radiology department database was searched to identify patient medical records from 2008 to 2013 that included the keywords "longus colli" and "deep flexors." Patients with fractures and ligament injuries were excluded. Patients with other obvious large soft-tissue injuries and nontraumatic conditions were also omitted. A total of 12 patients met the inclusion criterion of having an isolated or predominant injury to the LCM. Five patients had been involved in a motor vehicle accident, and seven patients had fallen. Eleven patients had undergone a CT examination before MRI was performed. RESULTS: No fractures were noted on CT. MRI examinations of the cervical spine were obtained for the following reasons: for increased prevertebral soft-tissue swelling noted on a CT scan plus neck pain (n = 6), for neck pain only (n = 4), or as part of a routine protocol for assessment of obtunded patients (n = 2). Eight of the 12 patients had isolated injury to the LCM. The remaining four patients also had minor injuries to the other neck muscles. The MR image showed swelling and T2 hyperintensity in the LCM and revealed free fluid in the prevertebral space. CONCLUSION: Isolated injury to the LCM may occur in neck injuries. The MRI findings indicating such injury include increased T2 signal, swelling of the muscle, and the presence of prevertebral fluid.

Skull fractures in pediatric patients on computerized tomogram: comparison between routing bone window images and 3D volume-rendered images.

Skull fracture is a common finding following head trauma. It has a prognostic significance and its presence points to severe trauma. Additionally, there is a greater possibility of detecting associated small underlying extra-axial hematomas and subtle injuries to the brain parenchyma. In pediatric patients, the presence of multiple open sutures often makes fracture evaluation challenging. In our experience, 3D volume (3DV)-rendered CT images complement routine axial bone window (RBW) images in detection and characterization of fractures. This is a multi-reader, multi-case, paired retrospective study to compare the sensitivity and specificity of RBW and 3DV images in detection of calvarial fractures in pediatric patients. A total of 60 cases (22 with fractures and 38 without) were analyzed. Two experienced neuroradiologists and a radiology trainee were the readers of the study. For all readers, the sensitivity was not statistically different between the RBW and the 3DV interpretations. For each reader, there was a statistically significant difference in the interpretation times between the RBW and the 3DV viewing formats. A greater number of sutural diastasis was identified on 3DV. We propose that 3DV images should be part of routine head trauma imaging, especially in the pediatric age group. It requires minimal post-processing time and no additional radiation. Furthermore, 3DV images help in reducing the interpretation time and also enhance the ability of the radiologist to characterize the calvarial fractures.

Correlation of apparent diffusion coefficient with Ki-67 proliferation index in grading meningioma.

OBJECTIVE: A noninvasive method to predict aggressiveness of high-grade meningiomas would be desirable because it would help anticipate tumor recurrence and improve tumor management and the treatment outcomes. The Ki-67 protein is a marker of tumor proliferation, and apparent diffusion coefficient (ADC) is related to tumor cellularity. Therefore, we sought to determine whether there is a statistically significant correlation between ADC and Ki-67 values in meningiomas and whether ADC values can differentiate various meningioma subtypes. MATERIALS AND METHODS: MRI examinations and histopathology of 68 surgically treated meningiomas were retrospectively reviewed. Mean ADC values were derived from diffusion imaging. Correlation coefficients were calculated for mean ADC and Ki-67 proliferation index values using linear regression. An independent unpaired Student t test was used to compare the ADC and Ki-67 proliferation index values from low-grade and more aggressive meningiomas. RESULTS: A statistically significant inverse correlation was found between ADC and Ki-67 proliferation index for low-grade and aggressive meningiomas (r(2) = -0.33, p = 0.0039). ADC values (± SD) of low-grade meningiomas (0.84 ± 0.14 × 10(-3) mm(2)/s) and aggressive (atypical or anaplastic) meningiomas (0.75 ± 0.03 × 10(-3) mm(2)/s) were significantly different (p = 0.0495). Using an ADC cutoff value of 0.70 × 10(-3) mm(2)/s, the sensitivity for diagnosing aggressive meningiomas was 29%, specificity was 94%, positive predictive value was 67%, and negative predictive value was 75%. CONCLUSION: ADC values correlate inversely with Ki-67 proliferation index and help differentiate low-grade from aggressive meningiomas.

Sinus pericranii with a hair collar sign.

Sinus pericranii is a rare vascular malformation in which the extracranial veins are connected to the intracranial venous circulation. It typically appears as a soft mass with a blue hue on the scalp. We describe a patient with sinus pericranii and associated hair collar sign.

Flow artifact in the anterior communicating artery resembling aneurysm on the time of flight MR angiogram.

BACKGROUND: Magnetic resonance angiogram (MRA) of the brain is a widely employed non-invasive test to diagnose aneurysms. However, its overall accuracy is less than digital subtraction angiography and is prone to give false-positive or false-negative results. False-negative results can be seen with hemorrhage, lipoma, dermoid, posterior lobe of the pituitary gland, and the flow artifacts. PURPOSE: To describe the findings associated with false aneurysms in the anterior communicating artery on the time of flight MRA and review the physical principles behind this artifact. MATERIAL AND METHODS: This short series comprises of four patients whose MRA showed suspicious aneurysms in the region of the anterior communicating artery (ACOM) on time of flight MRA. RESULTS: Two patients underwent catheter angiogram and the other two patients had computed tomography angiogram. None of these cases proved to have aneurysms and normal anterior communicating arteries were seen in all the patients. The findings on the MR angiograms were considered artifacts. All of the pseudoaneurysm had tapered appearance. CONCLUSION: MR angiogram can result in artifacts at ACOM which may be mistaken for aneurysm. Such pseudoaneurysms have characteristic appearance and should be followed up with non-invasive tests.

Role of noncontrast head CT in the assessment of vascular abnormalities in the emergency room.

Noncontrast CT of the head is a widely used noninvasive investigation for a variety of acute and chronic neurological conditions. Since CT head without contrast is usually the first and often the only investigation in the emergency room for many neurological symptoms, it is imperative to detect subtle vascular changes, which in many patients can be life-saving. The vascular abnormalities may present with increased density and/or size of the vessels, filling defects, and be associated with parenchymal and bony changes. In this article, we present examples of several vascular pathologies which can be identified on the noncontrast CT of the head, and learn imaging and interpretation techniques to help recognize what often are nebulous changes. While some of the findings are diagnostic by themselves and others subtle, any suspicious abnormality should be followed with dedicated vascular imaging such as CT/MR angiogram, venogram, or catheter angiogram for confirmation and better characterization.

Improved image quality and detection of small cerebral infarctions with diffusion-tensor trace imaging.

OBJECTIVE: The purpose of this study was to test a hypothesis that routinely performed diffusion-tensor trace imaging is of sufficient image quality and sensitivity for infarct detection to safely and routinely replace standard diffusion-weighted imaging (DWI) in the clinical setting. MATERIALS AND METHODS: Both routine DWI and 15-direction diffusion-tensor imaging (DTI) with parallel acquisition technique were obtained on all brain MRI studies from a single 1.5-T MRI scanner at a tertiary care referral center over a 1-year period, permitting direct comparison of the two different diffusion studies on the same patients (2537 studies, 365 infarct-positive studies). A subset of images was assessed for image quality and quantitatively for ability to detect brain infarctions. The total set of positive studies was reviewed qualitatively for ability to detect small cerebral infarctions. RESULTS: Fifteen-direction isotropic DWI (DTI trace images) with parallel acquisition technique resulted in consistently higher image quality with less distortion and higher image detail than routine DWI. Small infarcts were better seen, and in 12 cases, infarcts could only be seen on 15-direction isotropic diffusion-weighted images. The additional scanning time required for 15-direction isotropic DWI did not result in significantly increased motion-related reduction in image quality compared with standard DWI. CONCLUSION: Diffusion-tensor trace images obtained with parallel acquisition technique are of improved image quality and improved sensitivity for detection of small cerebral infarctions relative to standard DWI. If such DTI data are acquired, routine DWI can be omitted.

Vertebral artery ostial stenosis: prevalence by digital subtraction angiography, MR angiography, and CT angiography.

BACKGROUND AND PURPOSE: (1) To determine the prevalence of vertebral arterial ostial stenosis (VOS) as determined by the "gold standard" of digital subtraction angiography (DSA). (2) To learn the correlation between vertebral ostial stenosis and study indication. (3) To determine the ability of contrast-enhanced magnetic resonance angiography (CE MRA) and computed tomographic angiography (CTA) to reflect the true prevalence of vertebral ostial stenosis as determined by DSA. METHODS: Three hundred and twenty-nine patients who underwent DSA had recorded evaluation of 443 vertebral artery origins. Cases were categorized by patient age and study indication. Similar numbers of CTA and MRA studies were assessed. RESULTS: The prevalence of VOS in the study population was 5.4%. VOS was not observed in patients under 40 years of age, and was seen in 12.5% of patients over 70 years. CE MRA demonstrated decreased signal at the vertebral origins consistent with stenosis in 20% of patients. CTA estimated VOS at .8%, and yielded 7.3% of studies, which were nondiagnostic for VOS. CONCLUSION: The prevalence of VOS as determined by DSA is low and increases with patient age and correlates with factors such as anterior infarct (18.4%), posterior infarct (33.3%), carotid atherosclerosis (30.8%), and vertebrobasilar insufficiency (33%). Patients being evaluated for reasons less closely correlated with atherosclerotic disease, such as arteriovenous malformation (AVM) or hemorrhage showed a lower prevalence of VA stenosis (brain aneurysm or AVM 5/121, 4.1%, brain hemorrhage 5/153, 3.3%). Routine clinical MRA significantly overestimates VOS prevalence, and findings suggest that CTA underestimates the degree and prevalence of VOS.

MR imaging of acute cervical spinal ligamentous and soft tissue trauma.

The increasing availability of magnetic resonance imaging (MRI) and the high sensitivity of MRI for soft tissue injury are resulting in the increased use of MRI for the evaluation of acute trauma. As cervical spine injury can have a devastating consequence, MRI is being more commonly used to evaluate cervical spine injury in the acute setting, necessitating emergent interpretation by the on-call radiologist. Unless one is formally trained in a trauma center, the MRI findings of soft tissue and ligamentous cervical spine injury may not be fully appreciated. The goal of this pictorial review is to familiarize the reader with some of the more common soft tissue, vascular, and ligamentous injuries seen on MRI of the cervical spine in the emergent setting.