MR Vessel Wall Imaging for Atherosclerosis and Vasculitis.

Conventional imaging modalities, such as computed tomography angiography, MR angiography, transcranial Doppler ultrasonography, and digital subtraction angiography, are utilized in evaluating intraluminal or intravascular pathology of the intracranial vessels. Limitations of luminal imaging techniques can lead to inaccurate diagnosis, evaluation, and risk stratification, as many cerebrovascular pathologies contain an extrinsic vessel wall component. Furthermore, vessel wall imaging can provide information regarding extent, treatment response, and biopsy targets for vasculitis cases. Overall, while vessel wall imaging can provide robust data regarding intracranial pathologies, further prospective, multicenter studies are required to improve diagnostic application and accuracy.

Bivalirudin as a substitute for heparin in neurointervention for patients with heparin-induced thrombocytopenia.

OBJECTIVES: Heparin-induced thrombocytopenia is a known complication of heparin exposure with potentially life-threatening sequelae. Direct thrombin inhibitors can be substituted for heparin in patients with heparin-induced thrombocytopenia that require anticoagulation. However, the use of direct thrombin inhibitors as a substitute for heparin has not been widely reported in the neuroendovascular literature. MATERIALS AND METHODS: Here we report the first use of the direct thrombin inhibitor bivalirudin in a neuroendovascular procedure as a substitute for heparin in a patient with a ruptured pseudoaneurysm and heparin-induced thrombocytopenia, and review the literature on the use of bivalirudin and argatroban for such patients. RESULTS: Bivalirudin was safely and effectively used in the case reported, with no thrombotic or hemorrhagic complications. Our literature review revealed a paucity of studies on the use of heparin alternatives, including bivalirudin, in neuroendovascular procedures in patients with heparin-induced thrombocytopenia. CONCLUSIONS: Heparin-induced thrombocytopenia is an important iatrogenic disease process in patients undergoing neuroendovascular procedures, and developing protocols to diagnose and manage heparin-induced thrombocytopenia is important for healthcare systems. While further research needs to be done to establish the full range of anticoagulation options to substitute for heparin, our case indicates bivalirudin as a potential candidate.

Background Subtraction Angiography with Deep Learning Using Multi-frame Spatiotemporal Angiographic Input.

Catheter Digital Subtraction Angiography (DSA) is markedly degraded by all voluntary, respiratory, or cardiac motion artifact that occurs during the exam acquisition. Prior efforts directed toward improving DSA images with machine learning have focused on extracting vessels from individual, isolated 2D angiographic frames. In this work, we introduce improved 2D + t deep learning models that leverage the rich temporal information in angiographic timeseries. A total of 516 cerebral angiograms were collected with 8784 individual series. We utilized feature-based computer vision algorithms to separate the database into "motionless" and "motion-degraded" subsets. Motion measured from the "motion degraded" category was then used to create a realistic, but synthetic, motion-augmented dataset suitable for training 2D U-Net, 3D U-Net, SegResNet, and UNETR models. Quantitative results on a hold-out test set demonstrate that the 3D U-Net outperforms competing 2D U-Net architectures, with substantially reduced motion artifacts when compared to DSA. In comparison to single-frame 2D U-Net, the 3D U-Net utilizing 16 input frames achieves a reduced RMSE (35.77 ± 15.02 vs 23.14 ± 9.56, p < 0.0001; mean ± std dev) and an improved Multi-Scale SSIM (0.86 ± 0.08 vs 0.93 ± 0.05, p < 0.0001). The 3D U-Net also performs favorably in comparison to alternative convolutional and transformer-based architectures (U-Net RMSE 23.20 ± 7.55 vs SegResNet 23.99 ± 7.81, p < 0.0001, and UNETR 25.42 ± 7.79, p < 0.0001, mean ± std dev). These results demonstrate that multi-frame temporal information can boost performance of motion-resistant Background Subtraction Deep Learning algorithms, and we have presented a neuroangiography domain-specific synthetic affine motion augmentation pipeline that can be utilized to generate suitable datasets for supervised training of 3D (2d + t) architectures.

Evaluating outcome associations with race after mechanical thrombectomy: an analysis of the NVQI-QOD acute ischemic stroke registry.

BACKGROUND: Mechanical thrombectomy has become the standard of care for acute ischemic stroke due to large vessel occlusions. Racial differences in outcomes after mechanical thrombectomy for acute ischemic stroke have not been extensively studied. We evaluate the real-world evidence for differences between races in the outcomes of thrombectomy for large vessel occlusions using the NeuroVascular Quality Initiative-Quality Outcomes Database (NVQI-QOD). METHODS: Data from the NVQI-QOD acute ischemic stroke registry were analyzed and compared for racial differences in outcomes after mechanical thrombectomy in 4507 patients from 28 US centers (17 states) between January 2014 and April 2021. Race was dichotomized into non-Hispanic White (NHW, n=3649) and non-Hispanic Black (NHB, n=858). We performed 1:1 propensity score matching resulting in a subsample of matched groups (n=761 each for NHB and NHW) to compare study endpoints using Welch's two-sided t-tests and Χ2 test for continuous and categorical outcomes, respectively. RESULTS: Prior to matching, NHW and NHB patients significantly differed in age, comorbidities, medication use, smoking status, and presenting stroke severity. No significant difference in functional outcomes or mortality, at discharge or follow-up, were revealed. NHB patients had higher average postprocedure length of stay than NHW patients, which persisted following matching (11.2 vs 9.1 days, P=0.004). CONCLUSION: Evidence from the NVQI-QOD acute ischemic stroke registry showed that outcome metrics, such as modified Rankin Scale score and mortality, did not differ significantly between racial groups; however, disparity between NHW and NHB patients in postprocedure length of stay following mechanical thrombectomy was revealed.

Clot composition and recanalization outcomes in mechanical thrombectomy.

BACKGROUND: Mechanical thrombectomy (MT) has become standard for large vessel occlusions, but rates of complete recanalization are suboptimal. Previous reports correlated radiographic signs with clot composition and a better response to specific techniques. Therefore, understanding clot composition may allow improved outcomes. METHODS: Clinical, imaging, and clot data from patients enrolled in the STRIP Registry from September 2016 to September 2020 were analyzed. Samples were fixed in 10% phosphate-buffered formalin and stained with hematoxylin-eosin and Martius Scarlett Blue. Percent composition, richness, and gross appearance were evaluated. Outcome measures included the rate of first-pass effect (FPE, modified Thrombolysis in Cerebral Infarction 2c/3) and the number of passes. RESULTS: A total of 1430 patients of mean±SD age 68.4±13.5 years (median (IQR) baseline National Institutes of Health Stroke Scale score 17.2 (10.5-23), IV-tPA use 36%, stent-retrievers (SR) 27%, contact aspiration (CA) 27%, combined SR+CA 43%) were included. The median (IQR) number of passes was 1 (1-2). FPE was achieved in 39.3% of the cases. There was no association between percent histological composition or clot richness and FPE in the overall population. However, the combined technique resulted in lower FPE rates for red blood cell (RBC)-rich (P<0.0001), platelet-rich (P=0.003), and mixed (P<0.0001) clots. Fibrin-rich and platelet-rich clots required a higher number of passes than RBC-rich and mixed clots (median 2 and 1.5 vs 1, respectively; P=0.02). CA showed a trend towards a higher number of passes with fibrin-rich clots (2 vs 1; P=0.12). By gross appearance, mixed/heterogeneous clots had lower FPE rates than red and white clots. CONCLUSIONS: Despite the lack of correlation between clot histology and FPE, our study adds to the growing evidence supporting the notion that clot composition influences recanalization treatment strategy outcomes.

Unplanned readmission after carotid stenting versus endarterectomy: analysis of the United States Nationwide Readmissions Database.

BACKGROUND: Hospital readmissions are costly and reflect negatively on care delivered. OBJECTIVE: To have a better understanding of unplanned readmissions after carotid revascularization, which might help to prevent them. METHODS: The Nationwide Readmissions Database was used to determine rates and reasons for unplanned readmission following carotid endarterectomy (CEA) and carotid artery stenting (CAS). Trends were assessed by annual percent change, modified Poisson regression was used to estimate risk ratios (RR) for readmission, and propensity scores were used to match cohorts. RESULTS: Analysis yielded 522 040 asymptomatic and 55 485 symptomatic admissions for carotid revascularization between 2010 and 2015. Higher 30-day readmission rates were noted after CAS versus CEA in both symptomatic (9.1% vs 7.7%, p<0.001) and asymptomatic (6.8% vs 5.7%, p<0.001) patients. Readmission rates trended lower over time, significantly so for 90-day readmissions in symptomatic patients undergoing CEA. The most common cause for 30-day readmission was stroke in both symptomatic (5.5%) and asymptomatic (3.9%) patients. Factors associated with a higher risk of readmission included age over 80; male gender; Medicaid health insurance; and increases in severity of illness, mortality risk, and comorbidity indices. Analysis of matched cohorts showed that CAS had higher readmission than CEA (RR=1.14 (95% CI 1.06 to 1.22); p<0.001) only in asymptomatic patients. Adverse events during initial admission which predicted 30-day readmission included acute renal failure and acute respiratory failure in asymptomatic patients; hematoma and cardiac events were additional predictive adverse events in symptomatic patients. CONCLUSIONS: Readmission is not uncommon after carotid revascularization, occurs more often after CAS, and is predicted by baseline factors and by preventable adverse events at initial admission.

Correlation of von Willebrand factor and platelets with acute ischemic stroke etiology and revascularization outcome: an immunohistochemical study.

BACKGROUND: Platelets and von Willebrand factor (vWF) are key components of acute ischemic stroke (AIS) emboli. We aimed to investigate the CD42b (platelets)/vWF expression, its association with stroke etiology and the impact these components may have on the clinical/procedural parameters. METHODS: CD42b/vWF immunostaining was performed on 288 emboli collected as part of the multicenter STRIP Registry. CD42b/VWF expression and distribution were evaluated. Student's t-test and χ2 test were performed as appropriate. RESULTS: The mean CD42b and VWF content in clots was 44.3% and 21.9%, respectively. There was a positive correlation between platelets and vWF (r=0.64, p<0.001**). We found a significantly higher vWF level in the other determined etiology (p=0.016*) and cryptogenic (p=0.049*) groups compared with cardioembolic etiology. No significant difference in CD42b content was found across the etiology subtypes. CD42b/vWF patterns were significantly associated with stroke etiology (p=0.006*). The peripheral pattern was predominant in atherosclerotic clots (36.4%) while the clustering (patchy) pattern was significantly associated with cardioembolic and cryptogenic origin (66.7% and 49.8%, respectively). The clots corresponding to other determined etiology showed mainly a diffuse pattern (28.1%). Two types of platelets were distinguished within the CD42b-positive clusters in all emboli: vWF-positive platelets were observed at the center, surrounded by vWF-negative platelets. Thrombolysis correlated with a high platelet content (p=0.03*). vWF-poor and peripheral CD42b/vWF pattern correlated with first pass effect (p=0.03* and p=0.04*, respectively). CONCLUSIONS: The vWF level and CD42b/vWF distribution pattern in emboli were correlated with AIS etiology and revascularization outcome. Platelet content was associated with response to thrombolysis.

Middle meningeal artery embolization in pediatric patient.

PURPOSE: There is paucity of data in management of recurrent and expanding subdural hematomas (SDH) within the pediatric population, who are otherwise not surgical candidates. Middle meningeal artery (MMA) embolization has been utilized minimally in this population and here, we explore the utility of this procedure in a 15-month-old-child, along with review of the literature. METHODS: A case report of a 15-month-old child who underwent MMA embolization for recurrent and expanding SDH in the setting of anticoagulation for cardiac condition. A literature review of MMA embolization in pediatric patients was conducted. RESULTS: Initially stabilization of SDH was noted on serial imaging; however, recurrent hemorrhages were noted with subsequent boluses of antiplatelet and anticoagulating agents. There are only 5 total reported cases, included ours, of MMA embolization in pediatrics with an overall success rate of 80%. CONCLUSION: Treatment of chronic or recurrent subdural hematoma by MMA embolization in the pediatric population is understudied. Our case notes limitation of this procedure and impact on long-term success, specifically in patients with systemic illness and ongoing anticoagulation.

Holohemispheric Prostate Carcinoma Dural Metastasis Mimicking Subdural Hematoma: Case Report and Review of the Literature.

Prostate carcinomas are the most common malignancy to metastasize to the dura. These metastases can commonly mimic subdural hematomas and may similarly present with brain compression. The optimal management and outcomes after surgical management are not well characterized. We present a case of prostate carcinoma metastatic to the dura that was initially thought to be a large isodense subdural hematoma and was treated with surgical decompression. We also review the literature regarding prostate dural metastases mimicking subdural hematomas and discuss the relevant imaging findings, treatments, and outcomes. Dural metastasis should be considered when a patient with known metastatic prostate cancer presents with imaging evidence of a subdural mass.

Histological composition of retrieved emboli in acute ischemic stroke is independent of pre-thrombectomy alteplase use.

BACKGROUND AND PURPOSE: Given recent evidence suggesting the clot composition may be associated with revascularization outcomes and stroke etiology, clot composition research has been a topic of growing interest. It is currently unclear what effect, if any, pre-thrombectomy thrombolysis has on clot composition. Understanding this association is important as it is a potential confounding variable in clot composition research. We retrospectively evaluated the composition of retrieved clots from ischemic stroke patients who did and did not receive pre-treatment tPA to study the effect of tPA on clot composition. MATERIALS AND METHODS: Consecutive patients enrolled in the Stroke Thromboembolism Registry of Imaging and Pathology (STRIP) were included in this study. All patients underwent mechanical thrombectomy and retrieved clots were sent to a central core lab for processing. Histological analysis was performed using Martius Scarlett Blue (MSB) staining and area of the clot was also measured on the gross photos. Student's t test was used for continuous variables and chi-squared test for categorical variables. RESULTS: A total of 1430 patients were included in this study. Mean age was 68.4±13.5 years. Overall rate of TICI 2c/3 was 67%. A total of 517 patients received tPA (36%) and 913 patients did not (64%). Mean RBC density for the tPA group was 42.97±22.62% compared to 42.80±23.18% for the non-tPA group (P=0.89). Mean WBC density for the tPA group was 3.74±2.60% compared to 3.42±2.21% for the non-tPA group (P=0.012). Mean fibrin density for the tPA group was 26.52±15.81% compared to 26.53±15.34% for the non-tPA group (P=0.98). Mean platelet density for the tPA group was 26.22±18.60% compared to 26.55±19.47% for the non-tPA group (P=0.75). tPA group also had significantly smaller clot area compared to non-tPA group. CONCLUSIONS: Our study 1430 retrieved emboli and ischemic stroke patients shows no interaction between tPA administration and clot composition. These findings suggest that tPA does not result in any histological changes in clot composition.

Histological evaluation of acute ischemic stroke thrombi may indicate the occurrence of vessel wall injury during mechanical thrombectomy.

BACKGROUND: Several animal studies have demonstrated that mechanical thrombectomy (MT) for acute ischemic stroke (AIS) may cause vessel wall injury (VWI). However, the histological changes in human cerebral arteries following MT are difficult to determine. OBJECTIVE: To investigate the occurrence of VWI during MT by histological and immunohistochemical evaluation of AIS clots. METHODS: As part of the multicenter STRIP registry, 277 clots from 237 patients were analyzed using Martius Scarlett Blue stain and immunohistochemistry for CD34 (endothelial cells) and smooth muscle actin (smooth muscle cells). RESULTS: MT devices used were aspiration catheters (100 cases), stentriever (101 cases), and both (36 cases). VWI was found in 33/277 clots (12%). There was no significant correlation between VWI and MT device. The degree of damage varied from grade I (mild intimal damage, 24 clots), to grade II (relevant intimal and subintimal damage, 3 clots), and III (severe injury, 6 clots). VWI clots contained significantly more erythrocytes (p=0.006*) and less platelets/other (p=0.005*) than non-VWI clots suggesting soft thrombus material.Thrombolysis correlated with a lower rate of VWI (p=0.04*). VWI cases showed a significantly higher number of passes (2 [1-4] vs 1 [1-3], p=0.028*) and poorer recanalization outcome (p=0.01*) than cases without VWI. CONCLUSIONS: Histological markers of VWI were present in 12% of AIS thrombi, suggesting that VWI might be related to MT. VWI was associated with soft thrombus consistency, higher number of passes and poorer revascularization outcome. There was no significant correlation between VWI and MT device.

Quantification of clot spatial heterogeneity and its impact on thrombectomy.

BACKGROUND: Compositional and structural features of retrieved clots by thrombectomy can provide insight into improving the endovascular treatment of ischemic stroke. Currently, histological analysis is limited to quantification of compositions and qualitative description of the clot structure. We hypothesized that heterogeneous clots would be prone to poorer recanalization rates and performed a quantitative analysis to test this hypothesis. METHODS: We collected and did histology on clots retrieved by mechanical thrombectomy from 157 stroke cases (107 achieved first-pass effect (FPE) and 50 did not). Using an in-house algorithm, the scanned images were divided into grids (with sizes of 0.2, 0.3, 0.4, 0.5, and 0.6 mm) and the extent of non-uniformity of RBC distribution was computed using the proposed spatial heterogeneity index (SHI). Finally, we validated the clinical significance of clot heterogeneity using the Mann-Whitney test and an artificial neural network (ANN) model. RESULTS: For cases with FPE, SHI values were smaller (0.033 vs 0.039 for grid size of 0.4 mm, P=0.028) compared with those without. In comparison, the clot composition was not statistically different between those two groups. From the ANN model, clot heterogeneity was the most important factor, followed by fibrin content, thrombectomy techniques, red blood cell content, clot area, platelet content, etiology, and admission of intravenous tissue plasminogen activator (IV-tPA). No statistical difference of clot heterogeneity was found for different etiologies, thrombectomy techniques, and IV-tPA administration. CONCLUSIONS: Clot heterogeneity can affect the clot response to thrombectomy devices and is associated with lower FPE. SHI can be a useful metric to quantify clot heterogeneity.

Case Report: Anterior Spinal Cord Ischemia Following Embolization of Cerebellar Arteriovenous Malformation: An Illustrative Case and Review of Spinal Cord Vascular Anatomy.

Spinal cord ischemia (SCI) is a rare entity with high mortality and morbidity which can arise from causes such as atherosclerosis, aortic dissection or aneurysm, thromboembolic events or systemic hypotension, and is a potential complication of spinal surgery. Published literature contains very few reports of SCI as a complication of intracranial interventions, highlighting the uncommon nature of SCI in these circumstances. We report the occurrence of anterior SCI in a 69-year-old patient following successful embolization of a cerebellar arteriovenous malformation (AVM), marked by upper extremity weakness, lower extremity paraplegia, loss of bladder and bowel control, and hypercapnic respiratory failure requiring mechanical ventilation. Magnetic resonance imaging (MRI) demonstrated upper cervical diffusion restriction and T2/STIR hyperintensity. Unusually, SCI occurred in this case without intraprocedural catheter wedging or obvious flow limitation, prolonged procedure time, hypercoagulable state, or general hypotension. We review previous cases in the literature as well as spinal cord vascular anatomy, and discuss the possible etiologies of this complication. Spinal cord ischemia could be a very rare complication of neuroendovascular procedures even in the absence of warning signs and should be carefully evaluated in patients with suspected neurologic symptoms after such procedures.

Pipeline embolization of cerebral aneurysms in pediatric patients: combined systematic review of patient-level data and multicenter retrospective review.

OBJECTIVE: Cerebral aneurysms in the pediatric population are rare and optimal treatment strategies are not as well characterized as in adults. The Pipeline embolization device (PED) is an endoluminal flow diverter that is commonly used to treat aneurysms in adults, but experience with this device in children is limited. The authors sought to further characterize PED use and outcomes in this specific population by performing both a systematic review of patient-level data from studies reporting the use of the PED to treat pediatric aneurysms and a retrospective review of their experience. METHODS: A systematic review of the PubMed, Embase, and Scopus databases was performed to identify studies reporting the use of the PED in pediatric patients (age ≤ 18 years). Disaggregated data regarding demographics, aneurysm characteristics, treatment, and outcomes were collected. Retrospective data from the authors' two institutions were also included. RESULTS: Thirty studies comprising patient-level data on 43 pediatric patients with 47 aneurysms were identified. An additional 9 patients with 9 aneurysms were included from the authors' institutions for a total of 52 patients with 56 aneurysms. The mean patient age was 11.1 years. Presentations included aneurysm rupture (17.3%) and symptomatic mass effect (23.1%). Aneurysms were located in the anterior circulation in 55.4% of cases, and 73.2% were described as nonsaccular. Imaging follow-up was available for 89.3% with a mean follow-up of 13.3 months. Aneurysm occlusion was reported in 75%, with 1 case each (1.8%) demonstrating significant in-stent stenosis and parent vessel occlusion. Clinical follow-up was reported in 90.4% with a mean follow-up of 14.7 months. Good functional outcomes (modified Rankin Scale score of 0-1 or Glasgow Outcome Scale score of 5) were reported in 65.4% of the total population. Two major complications were reported, including 1 death. CONCLUSIONS: Despite substantial differences in aneurysm location and type between published pediatric and adult patient populations treated with the PED, the use of the PED in the pediatric population appears to be safe. While the short-term effectiveness is also similar to that of adults, additional studies are needed to further characterize the long-term outcomes and better define the use of this device in pediatric patients.

Per pass analysis of thrombus composition retrieved by mechanical thrombectomy.

BACKGROUND AND AIM: Mechanical thrombectomy (MT) for large vessel occlusion often requires multiple passes to retrieve the entire thrombus load. In this multi-institutional study we sought to examine the composition of thrombus fragments retrieved with each pass during MT. METHODS: Patients who required multiple passes during thrombectomy were included. Histopathological evaluation of thrombus fragments retrieved from each pass was performed using Martius Scarlet Blue staining and the composition of each thrombus component including RBC, fibrin and platelet was determined using image analysis software. RESULTS: 154 patients underwent MT and 868 passes was performed which resulted in 263 thrombus fragments retrieval. The analysis of thrombus components per pass showed higher RBC, lower fibrin and platelet composition in the pass 1 and 2 when compared to pass 3 and passes 4 or more combined (P values <0.05). There were no significant differences between thrombus fragments retrieved in pass 1 and pass 2 in terms of RBC, WBC, fibrin, and platelet composition (P values >0.05). Similarly, when each composition of thrombus fragments retrieved in pass 3 and passes 4 or more combined were compared with each other, no significant difference was noted (P values >0.05). CONCLUSION: Our findings confirm that thrombus fragments retrieved with each pass differed significantly in histological content. Fragments in the first passes were associated with lower fibrin and platelet composition compared to fragments retrieved in passes three and four or higher. Also, thrombus fragments retrieved after failed pass were associated with higher fibrin and platelet components.

Association between clot composition and stroke origin in mechanical thrombectomy patients: analysis of the Stroke Thromboembolism Registry of Imaging and Pathology.

BACKGROUND: We retrospectively evaluated the composition of retrieved clots from ischemic stroke patients to study the association between histological composition and stroke etiology METHODS: Consecutive patients enrolled in the Stroke Thromboembolism Registry of Imaging and Pathology (STRIP) were included in this study. All patients underwent mechanical thrombectomy and retrieved clots were sent to a central core lab for processing. Histological analysis was performed using martius scarlet blue (MSB) staining, and quantification for red blood cells (RBCs), white blood cells (WBCs), fibrin and platelets was performed using Orbit Image Software. A Wilcoxon test was used for continuous variables and χ2 test for categorical variables. RESULTS: 1350 patients were included in this study. The overall rate of Thrombolysis In Cerebral Infarction (TICI) 2c/3 was 68%. 501 patients received tissue plasminogen activator (tPA) (37%). 267 patients (20%) had a large artery atherosclerosis (LAA) source, 662 (49%) a cardioembolic (CE) source, 301 (22%) were cryptogenic, and the remainder had other identifiable sources including hypercoagulable state or dissection. LAA thrombi had a higher mean RBC density (46±23% vs 42±22%, p=0.01) and a lower platelet density (24±18% vs 27±18%, p=0.03) than CE thrombi. Clots from dissection patients had the highest mean RBC density (50±24%) while clots from patients with a hypercoagulable state had the lowest mean RBC density (26±21%). CONCLUSIONS: Our study found statistically significant but clinically insignificant differences between clots of CE and LAA etiologies. Future studies should emphasize molecular, proteomic and immunohistochemical characteristics to determine links between clot composition and etiology.

Direct percutaneous transcarotid approach for embolization of dural sinus malformation in a premature neonate: A case report.

CONCLUSIONS: Direct percutaneous TCAA can be a feasible access for intracranial interventions in neonates with LBW and unavailable UA access.

DeepCOVID-XR: An Artificial Intelligence Algorithm to Detect COVID-19 on Chest Radiographs Trained and Tested on a Large U.S. Clinical Data Set.

Background There are characteristic findings of coronavirus disease 2019 (COVID-19) on chest images. An artificial intelligence (AI) algorithm to detect COVID-19 on chest radiographs might be useful for triage or infection control within a hospital setting, but prior reports have been limited by small data sets, poor data quality, or both. Purpose To present DeepCOVID-XR, a deep learning AI algorithm to detect COVID-19 on chest radiographs, that was trained and tested on a large clinical data set. Materials and Methods DeepCOVID-XR is an ensemble of convolutional neural networks developed to detect COVID-19 on frontal chest radiographs, with reverse-transcription polymerase chain reaction test results as the reference standard. The algorithm was trained and validated on 14 788 images (4253 positive for COVID-19) from sites across the Northwestern Memorial Health Care System from February 2020 to April 2020 and was then tested on 2214 images (1192 positive for COVID-19) from a single hold-out institution. Performance of the algorithm was compared with interpretations from five experienced thoracic radiologists on 300 random test images using the McNemar test for sensitivity and specificity and the DeLong test for the area under the receiver operating characteristic curve (AUC). Results A total of 5853 patients (mean age, 58 years ± 19 [standard deviation]; 3101 women) were evaluated across data sets. For the entire test set, accuracy of DeepCOVID-XR was 83%, with an AUC of 0.90. For 300 random test images (134 positive for COVID-19), accuracy of DeepCOVID-XR was 82%, compared with that of individual radiologists (range, 76%-81%) and the consensus of all five radiologists (81%). DeepCOVID-XR had a significantly higher sensitivity (71%) than one radiologist (60%, P < .001) and significantly higher specificity (92%) than two radiologists (75%, P < .001; 84%, P = .009). AUC of DeepCOVID-XR was 0.88 compared with the consensus AUC of 0.85 (P = .13 for comparison). With consensus interpretation as the reference standard, the AUC of DeepCOVID-XR was 0.95 (95% CI: 0.92, 0.98). Conclusion DeepCOVID-XR, an artificial intelligence algorithm, detected coronavirus disease 2019 on chest radiographs with a performance similar to that of experienced thoracic radiologists in consensus. © RSNA, 2020 Supplemental material is available for this article. See also the editorial by van Ginneken in this issue.

Magnetization Transfer Ratio and Morphometrics of the Spinal Cord Associates with Surgical Recovery in Patients with Degenerative Cervical Myelopathy.

OBJECTIVES: We assessed the prognostic value of the preoperative magnetization transfer ratio (MTR) and morphometrics of the spinal cord in patients with degenerative cervical myelopathy (DCM) in a longitudinal cohort study. METHODS: Thirteen subjects with DCM underwent 3T magnetization transfer imaging. The MTR was calculated for the spinal cord regions and specific white matter tracts. Morphometric measures were extracted. Clinical (modified Japanese Orthopaedics Association [mJOA] and Nurick scale scores) and health-related quality of life scores were assessed before and after cervical decompression surgery. The association between the magnetic resonance imaging (MRI) metrics and postoperative recovery was assessed (Spearman's correlation). Receiver operating characteristics were used to assess the accuracy of MRI metrics in identifying ≥50% recovery in function. RESULTS: Preoperative anterior cord MTRs were associated with recovery in mJOA scores (ρ = 0.608; P = 0.036; area under the curve [AUC], 0.66). Preoperative lateral cord MTR correlated with the neck disability index (ρ = 0.699; P = 0.011) and pain interference scale (ρ = 0.732; P = 0.007). Preoperative rubrospinal tract MTR was associated with mJOA score recovery (ρ = 0.573; P = 0.041; AUC, 0.86). Preoperative corticospinal tract and reticulospinal MTRs were related to recovery in pain interference scores (ρ = 0.591; P = 0.033; and ρ = 0.583; P = 0.035, respectively). Eccentricity of the cord was associated with Nurick scores (ρ = 0.606; P = 0.028) and mJOA scores (ρ = 0.651; P = 0.025; AUC, 0.92). CONCLUSIONS: Preoperative MTR and eccentricity measurements of the spinal cord have prognostic value in assessing the response to surgery and recovery in patients with DCM. Advanced MRI and atlas-based postprocessing techniques can inform interventions and advance the healthcare received by patients with DCM.

Conservative Management and Natural History of Ruptured Basilar Perforator Artery Aneurysms: Two Cases and Literature Review.

BACKGROUND: Basilar perforator aneurysms are rare causes of subarachnoid hemorrhage and their natural history is poorly characterized. Although various treatment strategies have been reported, conservative management is an option that has been associated with a high likelihood of spontaneous resolution. CASE DESCRIPTION: Here we present 2 cases of subarachnoid hemorrhage, 1 diffuse and the other perimesencephalic, due to small ruptured basilar perforator artery aneurysms. These aneurysms were only identified after repeat angiography. Conservative management with serial imaging was pursued. Both patients did well clinically and repeat imaging demonstrated spontaneous resolution of the ruptured aneurysms. We also provide a literature review of ruptured basilar perforator aneurysms, showing a ~10% re-rupture rate within the early post-rupture period but otherwise a high rate of spontaneous resolution. CONCLUSIONS: Although basilar perforator aneurysms can re-rupture, there is also a high likelihood of spontaneous resolution. Given the challenges of treatment, conservative management is an option that can be considered.