Assessing the Impact of an Artificial Intelligence-Based Model for Intracranial Aneurysm Detection in CT Angiography on Patient Diagnosis and Outcomes (IDEAL Study)-a protocol for a multicenter, double-blinded randomized controlled trial

BACKGROUND

This multicenter, double-blinded, randomized controlled trial (RCT) aims to assess the impact of an artificial intelligence (AI)-based model on the efficacy of intracranial aneurysm detection in CT angiography (CTA) and its influence on patients' short-term and long-term outcomes.

METHODS

Study design: Prospective, multicenter, double-blinded RCT.

SETTINGS

The model was designed for the automatic detection of intracranial aneurysms from original CTA images.

PARTICIPANTS

Adult inpatients and outpatients who are scheduled for head CTA scanning. Randomization groups: (1) Experimental Group: Head CTA interpreted by radiologists with the assistance of the True-AI-integrated intracranial aneurysm diagnosis strategy (True-AI arm). (2) Control Group: Head CTA interpreted by radiologists with the assistance of the Sham-AI-integrated intracranial aneurysm diagnosis strategy (Sham-AI arm).

RANDOMIZATION

Block randomization, stratified by center, gender, and age group.

PRIMARY OUTCOMES

Coprimary outcomes of superiority in patient-level sensitivity and noninferiority in specificity for the True-AI arm to the Sham-AI arm in intracranial aneurysms.

SECONDARY OUTCOMES

Diagnostic performance for other intracranial lesions, detection rates, workload of CTA interpretation, resource utilization, treatment-related clinical events, aneurysm-related events, quality of life, and cost-effectiveness analysis.

BLINDING

Study participants and participating radiologists will be blinded to the intervention.

SAMPLE SIZE

Based on our pilot study, the patient-level sensitivity is assumed to be 0.65 for the Sham-AI arm and 0.75 for the True-AI arm, with specificities of 0.90 and 0.88, respectively. The prevalence of intracranial aneurysms for patients undergoing head CTA in the hospital is approximately 12%. To establish superiority in sensitivity and noninferiority in specificity with a margin of 5% using a one-sided α = 0.025 to ensure that the power of coprimary endpoint testing reached 0.80 and a 5% attrition rate, the sample size was determined to be 6450 in a 1:1 allocation to True-AI or Sham-AI arm.

DISCUSSION

The study will determine the precise impact of the AI system on the detection performance for intracranial aneurysms in a double-blinded design and following the real-world effects on patients' short-term and long-term outcomes.

TRIAL REGISTRATION

This trial has been registered with the NIH, U.S. National Library of Medicine at ClinicalTrials.gov, ID: NCT06118840 . Registered 11 November 2023.

Shifts in Diagnostic Testing for Headache in the Emergency Department, 2015 to 2021

Importance

Subarachnoid hemorrhage is typically diagnosed by noncontrast head computed tomography (CT); lumbar puncture is recommended if computed tomography is nondiagnostic, although CT cerebral angiography has been promoted as an alternative to lumbar puncture in this diagnostic pathway. The outcomes of this debate in practice have not been studied.

Objective

To determine whether CT cerebral angiography use has increased in lieu of lumbar puncture among emergency department (ED) patients with headache, with an increase in unruptured intracranial aneurysm detection.

Design, Setting, and Participants

This retrospective cohort study took place in 21 community EDs of an integrated health care system in Northern California between 2015 and 2021. Participants were adult (aged >17 years) health plan members with a chief concern of headache. Exclusions were prior diagnoses of subarachnoid hemorrhage, unruptured intracranial aneurysm, cerebral arteriovenous malformation, or cerebrospinal fluid shunt. Data were analyzed from October to November 2023.

Exposures

CT cerebral angiography and/or lumbar puncture during the ED encounter.

Main Outcomes and Measures

Primary and secondary outcomes were 14-day and 90-day unruptured intracranial aneurysm detection, respectively. Safety outcomes were missed diagnoses of subarachnoid hemorrhage or bacterial meningitis. The annual incidence of unruptured intracranial aneurysm detection was normalized to the incidence of subarachnoid hemorrhage (UIA:SAH ratio). Average annualized percentage changes were quantified using joinpoint regression analysis.

Results

Among 198 109 included ED encounters, the mean (SD) age was 47.5 (18.4) years; 140 001 patients (70.7%) were female; 29 035 (14.7%) were Black or African American, 59 896 (30.2%) were Hispanic or Latino, and 75 602 (38.2%) were White. Per year, CT cerebral angiography use increased (18.8%; 95% CI, 17.7% to 20.3%) and lumbar punctures decreased (-11.1%; 95% CI, -12.0% to -10.4%), with a corresponding increase in the 14-day UIA:SAH ratio (3.5%; 95% CI, 0.9% to 7.4%). Overall, computed tomography cerebral angiography use increased 6-fold relative to lumbar puncture, with a 33% increase in the detection of UIA. Results were similar at 90 days and robust to sensitivity analyses. Subarachnoid hemorrhage (1004 cases) and bacterial meningitis (118 cases) were misdiagnosed in 5% and 18% of cases, respectively, with no annual trends (P = .34; z1003 = .95 and P = .74; z117 = -.34, respectively).

Conclusions and Relevance

In this cohort study of ED patients with headache, increases in CT cerebral angiography use were associated with fewer lumbar punctures and higher detection of unruptured intracranial aneurysms, with no significant change in missed diagnoses of subarachnoid hemorrhage or bacterial meningitis. While this shift in diagnostic strategy appeared safe in the short-term, the long-term consequences remain unclear.

Detection of Intracranial Aneurysms Using Multiphase CT Angiography with a Deep Learning Model

RATIONALE AND OBJECTIVES

Determine the effect of a multiphase fusion deep-learning model with automatic phase selection in detection of intracranial aneurysm (IA) from computed tomography angiography (CTA) images.

MATERIALS AND METHODS

CTA images of intracranial arteries from patients at Ningbo First Hospital were retrospectively analyzed. Images were randomly classified as training data, internal validation data, or test data. CTA images from cases examined by digital subtraction angiography (DSA) were examined for independent validation. A deep-learning model was constructed by automatic phase selection of multiphase fusion, and compared to the single-phase algorithm to evaluate algorithm sensitivity.

RESULTS

We analyzed 1110 patients (1493 aneurysms) as training data, 139 patients (174 aneurysms) as internal validation data, and 134 patients (175 aneurysms) as test data. The sensitivity of the multiphase analysis of the internal validation data, test data, and independent validation data were greater than from the single-phase analysis. The recall of the multiphase selection was greater or equal to that of single-phase selection in the aneurysm position, shape, size, and rupture status. Use of the test data to determine the presence and absence of aneurysm rupture led to a recall from multiphase selection of 94.8% and 87.6% respectively; both of these values were greater than those from single-phase selection (89.6% and 79.4%).

CONCLUSION

A multiphase fusion deep learning model with automatic phase selection provided automated detection of IAs with high sensitivity.

Cost-effectiveness of repeat delayed imaging for spontaneous subarachnoid hemorrhage

BACKGROUND

In patients with intracranial aneurysm presenting with spontaneous subarachnoid hemorrhage (SAH), 15% of them could be missed by the initial diagnostic imaging. Repeat delayed imaging can help to identify previously undetected aneurysms, however, the cost-effectiveness of this strategy remains uncertain.

OBJECTIVE

The aim of this study is to assess the cost-effectiveness of repeat delayed imaging in patients with SAH who had a negative result during their initial imaging.

METHODS

A Markov model was developed to estimate the lifetime costs and quality-adjusted life-year (QALY) for patients who received or not received repeat delayed imaging. The analyses were conducted from a healthcare perspective, with costs reported in UK pounds and expressed in 2020 values. Extensive sensitivity analyses were performed to assess the robustness of the results.

RESULTS

The base case incremental cost-effectiveness ratio (ICER) of repeat delayed imaging is £9,314 per QALY compared to no-repeat delayed imaging. This ICER is below the National Institute for Health and Care Excellence (NICE) £20,000 per QALY willingness-to-pay threshold. At the NICE willingness-to-pay threshold of £20,000 per QALY, the probability that repeat delayed imaging is most cost-effective is 0.81. The results are sensitive to age, the utility of survived patients with a favorable outcome, the sensitivity of repeat delayed imaging, and the prevalence of aneurysm.

CONCLUSIONS

This study showed that, in the UK, it is cost-effective to provide repeat delayed imaging using computed tomographic angiography (CTA) for patients with SAH who had a negative result in their initial imaging.

2023 Guideline for the Management of Patients With Aneurysmal Subarachnoid Hemorrhage: A Guideline From the American Heart Association/American Stroke Association

AIM

The "2023 Guideline for the Management of Patients With Aneurysmal Subarachnoid Hemorrhage" replaces the 2012 "Guidelines for the Management of Aneurysmal Subarachnoid Hemorrhage." The 2023 guideline is intended to provide patient-centric recommendations for clinicians to prevent, diagnose, and manage patients with aneurysmal subarachnoid hemorrhage.

METHODS

A comprehensive search for literature published since the 2012 guideline, derived from research principally involving human subjects, published in English, and indexed in MEDLINE, PubMed, Cochrane Library, and other selected databases relevant to this guideline, was conducted between March 2022 and June 2022. In addition, the guideline writing group reviewed documents on related subject matter previously published by the American Heart Association. Newer studies published between July 2022 and November 2022 that affected recommendation content, Class of Recommendation, or Level of Evidence were included if appropriate. Structure: Aneurysmal subarachnoid hemorrhage is a significant global public health threat and a severely morbid and often deadly condition. The 2023 aneurysmal subarachnoid hemorrhage guideline provides recommendations based on current evidence for the treatment of these patients. The recommendations present an evidence-based approach to preventing, diagnosing, and managing patients with aneurysmal subarachnoid hemorrhage, with the intent to improve quality of care and align with patients' and their families' and caregivers' interests. Many recommendations from the previous aneurysmal subarachnoid hemorrhage guidelines have been updated with new evidence, and new recommendations have been created when supported by published data.

CT Angiography Radiomics Combining Traditional Risk Factors to Predict Brain Arteriovenous Malformation Rupture: a Machine Learning, Multicenter Study

This study aimed to develop a machine learning model for predicting brain arteriovenous malformation (bAVM) rupture using a combination of traditional risk factors and radiomics features. This multicenter retrospective study enrolled 586 patients with unruptured bAVMs from 2010 to 2020. All patients were grouped into the hemorrhage (n = 368) and non-hemorrhage (n = 218) groups. The bAVM nidus were segmented on CT angiography images using Slicer software, and radiomic features were extracted using Pyradiomics. The dataset included a training set and an independent testing set. The machine learning model was developed on the training set and validated on the testing set by merging numerous base estimators and a final estimator based on the stacking method. The area under the receiver operating characteristic (ROC) curve, precision, and the f1 score were evaluated to determine the performance of the model. A total of 1790 radiomics features and 8 traditional risk factors were contained in the original dataset, and 241 features remained for model training after L1 regularization filtering. The base estimator of the ensemble model was Logistic Regression, whereas the final estimator was Random Forest. In the training set, the area under the ROC curve of the model was 0.982 (0.967-0.996) and 0.893 (0.826-0.960) in the testing set. This study indicated that radiomics features are a valuable addition to traditional risk factors for predicting bAVM rupture. In the meantime, ensemble learning can effectively improve the performance of a prediction model.

Modern Imaging of Aneurysmal Subarachnoid Hemorrhage

In this review, we discuss the imaging of aneurysmal subarachnoid hemorrhage (SAH). We discuss emergency brain imaging, aneurysm detection techniques, and the management of CTA-negative SAH. We also review the concepts of cerebral vasospasm and delayed cerebral ischemia that occurs after aneurysm rupture and their impact on patient outcomes. These pathologies are distinct, and the use of multimodal imaging modalities is essential for prompt diagnosis and management to minimize morbidity from these conditions. Lastly, new advances in artificial intelligence and advanced imaging modalities such as PET and MR imaging scans have been shown to improve the detection of aneurysms and potentially predict outcomes early in the course of SAH.

LIVE-Net: Comprehensive 3D vessel extraction framework in CT angiography

The extraction of vessels from computed tomography angiography (CTA) is significant in diagnosing and evaluating vascular diseases. However, due to the anatomical complexity, wide intensity distribution, and small volume proportion of vessels, vessel extraction is laborious and time-consuming, and it is easy to lead to error-prone diagnostic results in clinical practice. This study proposes a novel comprehensive vessel extraction framework, called the Local Iterative-based Vessel Extraction Network (LIVE-Net), to achieve 3D vessel segmentation while tracking vessel centerlines. LIVE-Net contains dual dataflow pathways that work alternately: an iterative tracking network and a local segmentation network. The former can generate the fine-grain direction and radius prediction of a vascular patch by using the attention-embedded atrous pyramid network (aAPN), and the latter can achieve 3D vascular lumen segmentation by constructing the multi-order self-attention U-shape network (MOSA-UNet). LIVE-Net is trained and evaluated on two datasets: the MICCAI 2008 Coronary Artery Tracking Challenge (CAT08) dataset and head and neck CTA dataset from the clinic. Experimental results of both tracking and segmentation show that our proposed LIVE-Net exhibits superior performance compared with other state-of-the-art (SOTA) networks. In the CAT08 dataset, the tracked centerlines have an average overlap of 95.2%, overlap until first error of 91.2%, overlap with the clinically relevant vessels of 98.3%, and error distance inside of 0.21 mm. The corresponding tracking overlap metrics in the head and neck CTA dataset are 96.7%, 91.0%, and 99.8%, respectively. In addition, the results of the consistent experiment also show strong clinical correspondence. For the segmentation of bilateral carotid and vertebral arteries, our method can not only achieve better accuracy with an average dice similarity coefficient (DSC) of 90.03%, Intersection over Union (IoU) of 81.97%, and 95% Hausdorff distance (95%HD) of 3.42 mm , but higher efficiency with an average time of 67.25 s , even three times faster compared to some methods applied in full field view. Both the tracking and segmentation results prove the potential clinical utility of our network.

Delayed appearance of basilar trunk small atypical aneurysms in nontraumatic, initially angiogram-negative subarachnoid hemorrhage: A report of three patients

BACKGROUND

Repeat angiography will identify vascular pathology in approximately 10% of cases following angiogram-negative subarachnoid hemorrhage (anSAH), but small atypical aneurysms of the basilar artery are very uncommon.

OBJECTIVE

To report a case series of delayed appearance of nontraumatic basilar artery small atypical aneurysms.

METHODS

IRB approval was obtained for this retrospective case series and patient consent was waived.

RESULTS

Herein we report three cases of spontaneous anSAH, all of whom had a negative digital subtraction angiogram (DSA) on admission and all of whom had appearance of a small atypical aneurysms of the upper basilar trunk/apex on follow-up imaging (two during the initial admission and one in a delayed fashion). All three patients were ultimately treated with flow diversion (although one patient underwent attempted coiling that was abandoned due to inability to catheterize the aneurysm).

CONCLUSION

This report highlights the importance of a repeat DSA in cases of anSAH as well as the importance of scrutinizing the basilar trunk for these very small atypical aneurysms that may go unnoticed.

Early management of patients with aneurysmal subarachnoid hemorrhage in a hospital with neurosurgical/neuroendovascular facilities: a consensus and clinical recommendations of the Italian Society of Anesthesia and Intensive Care (SIAARTI)–Part 1

Background

Issues remain on the optimal management of subarachnoid hemorrhage (SAH) patients once they are admitted to the referring center, before and after the aneurysm treatment. To address these issues, we created a consensus of experts endorsed by the Italian Society of Anesthesia and Intensive Care (SIAARTI). In this manuscript, we aim to provide a list of experts’ recommendations regarding the early management of SAH patients from hospital admission, in a center with neurosurgical/neuro-endovascular facilities, until securing of the bleeding aneurysm.

Methods

A multidisciplinary consensus panel composed of 24 physicians selected for their established clinical and scientific expertise in the acute management of SAH patients with different background (anesthesia/intensive care, neurosurgery, and interventional neuroradiology) was created. A modified Delphi approach was adopted.

Results

Among 19 statements discussed. The consensus was reached on 18 strong recommendations. In one case, consensus could not be agreed upon and no recommendation was provided.

Conclusions

This consensus provides practical recommendations for the management of SAH patients in hospitals with neurosurgical/neuroendovascular facilities until aneurysm securing. It is intended to support clinician’s decision-making and not to mandate a standard of practice.

Evaluation of an automated intracranial aneurysm detection and rupture analysis approach using cascade detection and classification networks

Intracranial aneurysm is commonly found in human brains especially for the elderly, and its rupture accounts for a high rate of subarachnoid hemorrhages. However, it is time-consuming and requires special expertise to pinpoint small aneurysms from computed tomography angiography (CTA) images. Deep learning-based detection has helped improve much efficiency but false-positives still render difficulty to be ruled out. To study the feasibility of deep learning algorithms for aneurysm analysis in clinical applications, this paper proposes a pipeline for aneurysm detection, segmentation, and rupture classification and validates its performance using CTA images of 1508 subjects. A cascade aneurysm detection model is employed by first using a fine-tuned feature pyramid network (FPN) for candidate detection and then applying a dual-channel ResNet aneurysm classifier to further reduce false positives. Detected aneurysms are then segmented by applying a traditional 3D V-Net to their image patches. Radiomics features of aneurysms are extracted after detection and segmentation. The machine-learning-based and deep learning-based rupture classification can be used to distinguish ruptured and un-ruptured ones. Experimental results show that the dual-channel ResNet aneurysm classifier utilizing image and vesselness information helps boost sensitivity of detection compared to single image channel input. Overall, the proposed pipeline can achieve a sensitivity of 90 % for 1 false positive per image, and 95 % for 2 false positives per image. For rupture classification the area under curve (AUC) of 0.906 can be achieved for the testing dataset. The results suggest feasibility of the pipeline for potential clinical use to assist radiologists in aneurysm detection and classification of ruptured and un-ruptured aneurysms.

A companion to the preclinical common data elements and case report forms for in vivo rodent neuroimaging: A report of the TASK3-WG3 Neuroimaging Working Group of the ILAE/AES Joint Translational Task Force

The International League Against Epilepsy/American Epilepsy Society (ILAE/AES) Joint Translational Task Force established the TASK3 working groups to create common data elements (CDEs) for various aspects of preclinical epilepsy research studies, which could help improve the standardization of experimental designs. In this article, we discuss CDEs for neuroimaging data that are collected in rodent models of epilepsy, with a focus on adult rats and mice. We provide detailed CDE tables and case report forms (CRFs), and with this companion manuscript, we discuss the methodologies for several imaging modalities and the parameters that can be collected.

Imaging of Headache Attributed to Vascular Disorders

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

Artificial Intelligence-Enabled Medical Analysis for Intracranial Cerebral Hemorrhage Detection and Classification

Intracranial hemorrhage (ICH) becomes a crucial healthcare emergency, which requires earlier detection and accurate assessment. Owing to the increased death rate (around 40%), the earlier recognition and classification of disease using computed tomography (CT) images are necessary to ensure a favourable prediction and restrain the existence of neurologic deficits. Since the manual diagnosis approach is time-consuming, automated ICH detection and classification models using artificial intelligence (AI) models are required. With this motivation, this study introduces an AI-enabled medical analysis tool for ICH detection and classification (AIMA-ICHDC) using CT images. The proposed AIMA-ICHDC technique aims at identifying the presence of ICH and identifying the different grades. In addition, the AIMA-ICHDC technique involves the design of glowworm swarm optimization with fuzzy entropy clustering (GSO-FEC) technique for the segmentation process. Besides, the VGG-19 model was executed for generating a collection of feature vectors and the optimal mixed-kernel-based extreme learning machine (OMKELM) model is utilized as a classifier. To optimally select the weight parameter of the MKELM technique, the coyote optimization algorithm (COA) was utilized. A wide range of simulation analyses are carried out under varying aspects. As part of the AIMA-ICHDC method, ICH can be detected and graded using a single sample. For segmentation, the AIMA-ICHDC technique uses the GSO-FEC method, which is the design of glowworm swarm optimization (GSO). The comparative outcomes highlighted the betterment of the AIMA-ICHDC technique compared to the recent state-of-the-art ICH classification approaches in terms of several measures.

Artificial Intelligence Applications in Intracranial Aneurysm: Achievements, Challenges and Opportunities

Intracranial aneurysms present in about 3% of the general population and the number of detected aneurysms is continuously rising with the advances in imaging techniques. Intracranial aneurysm rupture carries a high risk of death or permanent disabilities; therefore assessment of the intracranial aneurysm along the entire course is of great clinical importance. Given the outstanding performance of artificial intelligence (AI) in image-based tasks, many AI-based applications have emerged in recent years for the assessment of intracranial aneurysms. In this review we will summarize the state-of-the-art of AI applications in intracranial aneurysms, emphasizing the achievements, and exploring the challenges. We will also discuss the future prospects and potential opportunities. This article provides an updated view of the AI applications in intracranial aneurysms and may act as a basis for guiding the related future works.

Cost-Effective Analysis of Different Diagnostic Strategies in Screening for Aneurysms After Spontaneous Subarachnoid Hemorrhage

PURPOSE

With an increasing number of patients being treated by coiling for aneurysms, using computed tomographic angiography (CTA) or magnetic resonance angiography (MRA) as an initial screening test becomes less important because digital subtraction angiography (DSA) is the first step of endovascular treatment procedure. The objective of this study is to investigate whether CTA or MRA remained to be the optimal screening strategy in patients with spontaneous subarachnoid hemorrhage (SAH).

METHODS

CTA, MRA, and DSA strategy were evaluated in a decision tree model created with TreeAge Pro Suite 2011. Input parameters were derived from published literature and our institutional database. Base case and sensitivity analyses were conducted to assess the cost-effectiveness of each strategy. A Monte Carlo simulation was performed with all parameters ranged among their distributions to evaluate the validation of results.

RESULTS

The base case scenario showed that MRA was the most cost-effective strategy. Using a willingness-to-pay threshold of ¥70,892/quality adjusted life year, MRA remained to be most cost-effective when its sensitivity is >0.907. DSA was not cost-effective compared to CTA or MRA unless over 91.56% of patients were treated by coiling. The Monte Carlo simulation reported DSA not to be a cost-effective strategy at willingness-to-pay of ¥70,892 in 99.99% of the iterations.

CONCLUSION

DSA is not cost-effective compared to CTA or MRA and should not be used as the initial diagnostic tool for spontaneous SAH.

Leukocytosis and C-Reactive Protein May Predict Development of Secondary Cerebral Vasospasm in Patients with Aneurysmal Subarachnoid Hemorrhage

Background and Objectives: Secondary cerebral vasospasm (CV) with subsequent delayed cerebral ischemia (DCI) after aneurysmal subarachnoid hemorrhage (aSAH) remains an unpredictable pathology. The aim of this retrospective study was to investigate the association between inflammatory parameters, white blood cell (WBC) count, and C-reactive protein plasma levels (CRP) and the occurrence of secondary CV in patients with aSAH. Materials and Methods: The medical records of 201 Intensive Care Unit patients in Riga East University Hospital with aSAH were retrospectively reviewed in a 24-month period. WBC count and CRP values were observed at admission to the hospital and on the third day. According to the inclusion criteria, 117 (48 males) participants were enrolled for further analysis, with average age of 56 ± 15 years (mean ± SD). In total, secondary CV was diagnosed in 21.4% of cases, and DCI in 22.4% of cases. The patients were classified into three groups: SAH-CV group (n = 25), SAH-DCI group (n = 12), and SAH or control group (n = 80), for comparative analysis. Results: We found that SAH-CV patients demonstrated notably higher inflammatory parameters compared to controls: WBC 13.2 ± 3.3 × 10⁹/L vs. 11.2 ± 3.7 × 10⁹/L; p = 0.01 and CRP median 9.3 mg/L vs. 1.9 mg/L; p < 0.001, respectively. We found that the odds of developing CV increased by 5% for each CRP increase of 1 mg/L at admission (OR, 1.05; CI, 1.014–1.087; p = 0.006). Concomitantly, the odds increased by 16% for every rise in WBC count of 1 × 10⁹/L (OR, 1.16; CI, 1.02–1.32; p = 0.02). WBC count was associated with the occurrence of CV with 96% sensitivity and 40% specificity, with a cut off level of 10.015 × 10⁹/L and AUC 0.683; p = 0.006. CRP displayed 54% sensitivity and 90% specificity with a cut off value of 8.9 mg/L and AUC 0.751; p < 0.001. Moreover, higher values of inflammatory parameters at admission correlated with a longer stay in ICU (r = 0.3, p = 0.002 for WBC count and r = 0.305, p = 0.002 for CRP values), and poor outcome (death) was significantly associated with higher CRP values at admission and on the third day (16.1. vs. 2.2. and 57.4. vs. 11.1, p < 0.001, respectively). Higher mortality was detected in SAH-CV patients (32%) compared to controls (6.3%; p < 0.001). Conclusions: Inflammatory parameters such as WBC count and CRP values at admission might be helpful to predict the development of secondary CV.

Multiplanar CT evaluation of aneurysm rupture signs in subarachnoid hemorrhage

PURPOSE

In subarachnoid hemorrhage, noncontrast CT features are used to guide the localization of ruptured aneurysms on CT angiography and DSA. Multiplanar CT may improve the localization of aneurysm rupture sites over axial plane CT alone.

METHODS

Multiplanar CT in three orthogonal planes was used to evaluate 94 cases of SAH. Two investigators independently evaluated each imaging plane for focal thick SAH with mass effect, intracerebral hemorrhage, focal edema, filling defect, subdural hemorrhage, and dominant intraventricular hemorrhage. Also, rupture site was qualitatively identified by combining these variables in each plane and combination of three planes. DSA served as the gold standard to locate the rupture sites.

RESULTS

Interobserver agreement was k 0.60 to 0.79 for axial, k 0.43 to 0.86 for coronal and k 0.43 to 0.74 for sagittal planes. Good to substantial agreement was observed for the localization of rupture site in three planes (focal SAH with mass effect - k 0.78 to 0.85; filling defect - k 0.95 to 1.0; intracerebral hemorrhage - k 1.0; focal edema k 1.0; subdural hemorrhage - k 0.61 to 0.83). Dominant intraventricular hemorrhage revealed significant association with DSA to locate ruptured aneurysms (Fisher's exact test - Pr <  = P (< 0.001)). With non-missing data, frequency of correct ratings to locate rupture site was 66/67 (99%) in axial plane, 59/66 (89%) in coronal plane, 64/67 (96%) in sagittal plane and 77/77 (100%) in combined 3 planes.

CONCLUSIONS

Multiplanar CT head is more successful than axial plane CT alone for the localization of aneurysm rupture sites in SAH.

Radiological investigations in non-aneurysmal subarachnoid haemorrhage: A 5-year review

•

All patients with a Fisher grade 2 bleed and a negative CT angiogram had catheter angiography negative for any abnormality.

•

Neuroradiologists identified vascular abnormalities not reported by district general hospitals.

•

Follow-up MRI may be a useful adjunct in subarachnoid haemorrhage.

ACR Appropriateness Criteria® Cerebrovascular Diseases-Aneurysm, Vascular Malformation, and Subarachnoid Hemorrhage

Cerebrovascular disease is a broad topic. This document focuses on the imaging recommendations for the varied clinical scenarios involving intracranial aneurysms, vascular malformations, and vasculitis, which all carry high risk of morbidity and mortality. Additional imaging recommendations regarding complications of these conditions, including subarachnoid hemorrhage and vasospasm, are also covered. While each variant presentation has unique imaging recommendations, the major focus of this document is neurovascular imaging techniques. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

Artificial Intelligence Trained by Deep Learning Can Improve Computed Tomography Diagnosis of Nontraumatic Subarachnoid Hemorrhage by Nonspecialists

Subarachnoid hemorrhage (SAH) is a serious cerebrovascular disease with a high mortality rate and is known as a disease that is hard to diagnose because it may be overlooked by noncontrast computed tomography (NCCT) examinations that are most frequently used for diagnosis. To create a system preventing this oversight of SAH, we trained artificial intelligence (AI) with NCCT images obtained from 419 patients with nontraumatic SAH and 338 healthy subjects and created an AI system capable of diagnosing the presence and location of SAH. Then, we conducted experiments in which five neurosurgery specialists, five nonspecialists, and the AI system interpreted NCCT images obtained from 135 patients with SAH and 196 normal subjects. The AI system was capable of performing a diagnosis of SAH with equal accuracy to that of five neurosurgery specialists, and the accuracy was higher than that of nonspecialists. Furthermore, the diagnostic accuracy of four out of five nonspecialists improved by interpreting NCCT images using the diagnostic results of the AI system as a reference, and the number of oversight cases was significantly reduced by the support of the AI system. This is the first report demonstrating that an AI system improved the diagnostic accuracy of SAH by nonspecialists.

Computed Tomography Angiography versus Digital Subtraction Angiography as a Primary Diagnostic Tool in Nontraumatic Subarachnoid Hemorrhage: Cost-Effectiveness Analysis Study

BACKGROUND

Digital subtraction angiography (DSA) and computed tomographic angiography (CTA) are used to identify the cause of nontraumatic subarachnoid hemorrhage (SAH). There is no consensus on which to choose as the first diagnostic tool. We aimed to compare the cost-effectiveness of CTA versus DSA as a primary tool for identifying the cause of nontraumatic SAH.

METHODS

A decision analysis model was built to simulate patients undergoing DSA or CTA as a primary diagnostic tool for the cause of nontraumatic SAH. The input data for the study were extracted from literature. Probabilistic and deterministic sensitivity analyses were performed to evaluate the robustness of the model.

RESULTS

In the base case calculation, it cost $1261.82 less and yielded 0.0001 quality-adjusted life year (QALY) when DSA was used as a primary diagnostic imaging tool for nontraumatic SAH. Choosing DSA as a primary tool was cost-effective in more than 65% of iterations in probabilistic sensitivity analysis. Deterministic sensitivity analyses show when the probability of using endovascular treatment is >47.2%, choosing DSA is more cost-effective; otherwise, CTA is more optimal. CTA is more cost-effective when the cost for DSA >2.6 × CTA + $600.

CONCLUSIONS

Based on current literature and our model DSA as a primary diagnostic tool for the cause of nontraumatic SAH is more cost-effective. However, in clinical practice physicians can choose either DSA or CTA according to the scale of endovascular procedures used in their center, as well as the cost correlation between CTA and DSA, which varies among institutions.

Subarachnoid Hemorrhage of Unknown Cause: Distribution and Role of Imaging

Subarachnoid hemorrhage of unknown cause represents approximately 10% to 15% of nontraumatic subarachnoid hemorrhages. The key factors in determining the management strategy for a presumed nonaneurysmal subarachnoid hemorrhage are the distribution, location, and amount of subarachnoid blood. Hemorrhage distribution on computed tomography can be categorized as follows: perimesencephalic, diffuse, sulcal, and primary intraventricular. The extent of the workup required in determining the cause of hemorrhage depends on the distribution of blood. The authors review the potential causes, differential diagnoses, and acute and long-term follow-up strategies in patients with subarachnoid hemorrhage of unknown cause.

Construction of competitive endogenous RNA network reveals regulatory role of long non-coding RNAs in intracranial aneurysm

BACKGROUND

Rupture of intracranial aneurysm (IA) is the main cause of devastating subarachnoid hemorrhage, which urges our understanding of the pathogenesis and regulatory mechanisms of IA. However, the regulatory roles of long non-coding RNAs (lncRNAs) in IA is less known.

RESULTS

We processed the raw SRR files of 12 superficial temporal artery (STA) samples and 6 IA samples to count files. Then the differentially expressed (DE) mRNAs, miRNAs, and lncRNAs between STAs and IAs were identified. The enrichment analyses were performed using DEmRNAs. Next, a lncRNA-miRNA-mRNA regulatory network was constructed using integrated bioinformatics analysis. In summary, 341 DElncRNAs, 234 DEmiRNAs, and 2914 DEmRNAs between the STA and IA. The lncRNA-miRNA-mRNA regulatory network of IA contains 91 nodes and 146 edges. The subnetwork of hub lncRNA PVT1 was extracted. The expression level of PVT1 was positively correlated with a majority of the mRNAs in its subnetwork. Moreover, we found that several mRNAs (CCND1, HIF1A, E2F1, CDKN1A, VEGFA, COL1A1 and COL5A2) in the PVT1 subnetwork served as essential components in the PI3K-Akt signaling pathway, and that some of the non-coding RNAs (ncRNAs) (PVT1, HOTAIR, hsa-miR-17, hsa-miR-142, hsa-miR-383 and hsa-miR-193b) interacted with these mRNAs.

CONCLUSION

Our annotations noting ncRNA's role in the pathway may uncover novel regulatory mechanisms of ncRNAs and mRNAs in IA. These findings provide significant insights into the lncRNA regulatory network in IA.

CT Angiography-Based Radiomics for Classification of Intracranial Aneurysm Rupture

Background: Intracranial aneurysm rupture is a devastating medical event with a high morbidity and mortality rate. Thus, timely detection and management are critical. The present study aimed to identify the aneurysm radiomics features associated with rupture and to build and evaluate a radiomics classification model of aneurysm rupture. Methods: Radiomics analysis was applied to CT angiography (CTA) images of 393 patients [152 (38.7%) with ruptured aneurysms]. Patients were divided at a ratio of 7:3 into retrospective training (n = 274) and prospective test (n = 119) cohorts. A total of 1,229 radiomics features were automatically calculated from each aneurysm. The feature number was systematically reduced, and the most important classifying features were selected. A logistic regression model was constructed using the selected features and evaluated on training and test cohorts. Radiomics score (Rad-score) was calculated for each patient and compared between ruptured and unruptured aneurysms. Results: Nine radiomics features were selected from the CTA images and used to build the logistic regression model. The radiomics model has shown good performance in the classification of the aneurysm rupture on training and test cohorts [area under the receiver operating characteristic curve: 0.92 [95% confidence interval CI: 0.89-0.95] and 0.86 [95% CI: 0.80-0.93], respectively, p < 0.001]. Rad-score showed statistically significant differences between ruptured and unruptured aneurysms (median, 2.50 vs. -1.60 and 2.35 vs. -1.01 on training and test cohorts, respectively, p < 0.001). Conclusion: The results indicated the potential of aneurysm radiomics features for automatic classification of aneurysm rupture on CTA images.

The role of imaging in the management of non-traumatic subarachnoid hemorrhage: a practical review

The purpose of this review is to understand the role of imaging in the diagnosis and management of non-traumatic subarachnoid hemorrhage (SAH). SAH is a life-threatening emergency and a relatively common entity, the most common etiology being ruptured aneurysms. Multiple conundrums exist in literature at various steps of its imaging workup: diagnosis, management, and follow-up. We target our review to highlight the most effective practice and suggest efficient workup plans based on literature search, and describe in detail the clinical diagnostic and prognostic scales, role of CT scan, lumbar puncture, and MR, including angiography in the diagnosis and workup of SAH and its complications, and try to simplify the conundrums. Practical knowledge of imaging workup of SAH can help guide correct management of these patients, so as to reduce morbidity and mortality without resource overutilization.

Toward human intervention-free clinical diagnosis of intracranial aneurysm via deep neural network

Intracranial aneurysm (IA) is an enormous threat to human health, which often results in nontraumatic subarachnoid hemorrhage or dismal prognosis. Diagnosing IAs on commonly used computed tomographic angiography (CTA) examinations remains laborious and time consuming, leading to error-prone results in clinical practice, especially for small targets. In this study, we propose a fully automatic deep-learning model for IA segmentation that can be applied to CTA images. Our model, called Global Localization-based IA Network (GLIA-Net), can incorporate the global localization prior and generates the fine-grain three-dimensional segmentation. GLIA-Net is trained and evaluated on a big internal dataset (1,338 scans from six institutions) and two external datasets. Evaluations show that our model exhibits good tolerance to different settings and achieves superior performance to other models. A clinical experiment further demonstrates the clinical utility of our technique, which helps radiologists in the diagnosis of IAs.

•

GLIA-Net is a deep learning method for the clinical diagnosis of IAs

•

It can be applied directly to CTA images without any laborious preprocessing

•

A clinical study demonstrates its effectiveness in assisting diagnosis

•

An IA dataset of 1,338 CTA cases from six institutions is publicly released

Intracranial aneurysms (IAs) are enormous threats to human health with a prevalence of approximately 4%. The rupture of IAs usually causes death or severe damage to the patients. To enhance the clinical diagnosis of IAs, we present a deep learning model (GLIA-Net) for IA detection and segmentation without laborious human intervention, which achieves superior diagnostic performance validated by quantitative evaluations as well as a sophisticated clinical study. We anticipate that the publicly released data and the artificial intelligence technique would help to transform the clinical diagnostics and precision treatments of cerebrovascular diseases. They may also revolutionize the landscape of healthcare and biomedical research in the future.

Noncontrast Magnetic Resonance Angiography in the Era of Nephrogenic Systemic Fibrosis and Gadolinium Deposition

ABSTRACT

Gadolinium-based contrast agents for clinical magnetic resonance imaging are overall safe. However, the discovery of nephrogenic systemic fibrosis in patients with severe renal impairment and gadolinium deposition in patients receiving contrast have generated developments in contrast-free imaging of the vasculature, that is, noncontrast magnetic resonance angiography. This article presents an update on noncontrast magnetic resonance angiography techniques, with comparison to other imaging alternatives. Potential benefits and challenges to implementation, and evidence to date for various clinical applications are discussed.

Neurovascular disease, diagnosis, and therapy: Subarachnoid hemorrhage and cerebral vasospasm

The worldwide incidence of spontaneous subarachnoid hemorrhage is about 6.1 per 100,000 cases per year (Etminan et al., 2019). Eighty-five percent of cases are due to intracranial aneurysms. The mean age of those affected is 55 years, and two-thirds of the patients are female. The prognosis is related mainly to the neurologic condition after the subarachnoid hemorrhage and the age of the patient. Overall, 15% of patients die before reaching the hospital, another 20% die within 30 days, and overall 75% are dead or remain disabled. Case fatality has declined by 17% over the last 3 decades. Despite the improvement in outcome probably due to improved diagnosis, early aneurysm repair, administration of nimodipine, and advanced intensive care support, the outcome is not very good. Even among survivors, 75% have permanent cognitive deficits, mood disorders, fatigue, inability to return to work, and executive dysfunction and are often unable to return to their premorbid level of functioning. The key diagnostic test is computed tomography, and the treatments that are most strongly supported by scientific evidence are to undertake aneurysm repair in a timely fashion by endovascular coiling rather than neurosurgical clipping when feasible and to administer enteral nimodipine. The most common complications are aneurysm rebleeding, hydrocephalus, delayed cerebral ischemia, and medical complications (fever, anemia, and hyperglycemia). Management also probably is optimized by neurologic intensive care units and multidisciplinary teams.

Deep Learning for Detecting Cerebral Aneurysms with CT Angiography

Background Cerebral aneurysm detection is a challenging task. Deep learning may become a supportive tool for more accurate interpretation. Purpose To develop a highly sensitive deep learning-based algorithm that assists in the detection of cerebral aneurysms on CT angiography images. Materials and Methods Head CT angiography images were retrospectively retrieved from two hospital databases acquired across four different scanners between January 2015 and June 2019. The data were divided into training and validation sets; 400 additional independent CT angiograms acquired between July and December 2019 were used for external validation. A deep learning-based algorithm was constructed and assessed. Both internal and external validation were performed. Jackknife alternative free-response receiver operating characteristic analysis was performed. Results A total of 1068 patients (mean age, 57 years ± 11 [standard deviation]; 660 women) were evaluated for a total of 1068 CT angiograms encompassing 1337 cerebral aneurysms. Of these, 534 CT angiograms (688 aneurysms) were assigned to the training set, and the remaining 534 CT angiograms (649 aneurysms) constituted the validation set. The sensitivity of the proposed algorithm for detecting cerebral aneurysms was 97.5% (633 of 649; 95% CI: 96.0, 98.6). Moreover, eight new aneurysms that had been overlooked in the initial reports were detected (1.2%, eight of 649). With the aid of the algorithm, the overall performance of radiologists in terms of area under the weighted alternative free-response receiver operating characteristic curve was higher by 0.01 (95% CI: 0.00, 0.03). Conclusion The proposed deep learning algorithm assisted radiologists in detecting cerebral aneurysms on CT angiography images, resulting in a higher detection rate. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Kallmes and Erickson in this issue.

Aneurysmal Subarachnoid Hemorrhage: the Last Decade

Aneurysmal subarachnoid hemorrhage (SAH) affects six to nine people per 100,000 per year, has a 35% mortality, and leaves many with lasting disabilities, often related to cognitive dysfunction. Clinical decision rules and more sensitive computed tomography (CT) have made the diagnosis of SAH easier, but physicians must maintain a high index of suspicion. The management of these patients is based on a limited number of randomized clinical trials (RCTs). Early repair of the ruptured aneurysm by endovascular coiling or neurosurgical clipping is essential, and coiling is superior to clipping in cases amenable to both treatments. Aneurysm repair prevents rebleeding, leaving the most important prognostic factors for outcome early brain injury from the hemorrhage, which is reflected in the neurologic condition of the patient, and delayed cerebral ischemia (DCI). Observational studies suggest outcomes are better when patients are managed in specialized neurologic intensive care units with inter- or multidisciplinary clinical groups. Medical management aims to minimize early brain injury, cerebral edema, hydrocephalus, increased intracranial pressure (ICP), and medical complications. Management then focuses on preventing, detecting, and treating DCI. Nimodipine is the only pharmacologic treatment that is approved for SAH in most countries, as no other intervention has demonstrated efficacy. In fact, much of SAH management is derived from studies in other patient populations. Therefore, further study of complications, including DCI and other medical complications, is needed to optimize outcomes for this fragile patient population.

AH, R. PA. The Yield of Repeat Angiography in Angiography-Negative Spontaneous Subarachnoid Hemorrhage

Objective  Despite the technological advancement in imaging, digital subtraction angiography (DSA) remains gold standard imaging modality for spontaneous subarachnoid hemorrhage (SAH). But even after DSA, around 15% of SAH remains elusive for the cause of the bleed. This is an institutional review to solve the mystery, “when is second DSA really indicated?”

Methods  In a retrospective review from January 2015 to December 2017, we evaluated cases of spontaneous SAH with initial negative DSA with repeat DSA after 6 weeks to rule out vascular abnormality. The spontaneous SAH was confirmed on noncontrast computed tomography (NCCT) and divided into two groups of perimesencephalic SAH (PM-SAH) or nonperimesencephalic SAH (nPM-SAH). The outcome was assessed by a modified Rankin’s score (mRS) at 6 months postictus.

Results  During the study period, we had 119 cases of initial negative DSA and 98 cases (82.3%) underwent repeat DSA after 6 weeks interval. A total of 53 cases (54.1%) had PM-SAH and 45 cases (45.9%) had nPM-SAH. Repeat DSA after 6 weeks showed no vascular abnormality in 53 cases of PM-SAH and in 2 (4.4%) out of 45 cases of nPM-SAH. At 6 months postictus, all cases of PM-SAH and 93% of nPM-SAH had mRS of 0.

Conclusion  We recommend, a repeat DSA is definitely not required in PM-SAH, but it should be done for all cases of nPM-SAH, before labeling them as nonaneurysmal SAH. Although the overall outcome for nonaneurysmal spontaneous SAH is better than aneurysmal SAH, nPM-SAH has poorer eventual outcome compared to PM-SAH.

Comprehensive review of imaging of intracranial aneurysms and angiographically negative subarachnoid hemorrhage

Intracranial aneurysms confer the risk of subarachnoid hemorrhage (SAH), a potentially devastating condition, though most aneurysms will remain asymptomatic for the lifetime of the patient. Imaging is critical to all stages of patient care for those who harbor an unruptured intracranial aneurysm (UIA), including to establish the diagnosis, to determine therapeutic options, to undertake surveillance in patients who elect not to undergo treatment or whose aneurysm(s) portends such a low risk that treatment is not indicated, and to perform follow-up after treatment. Neuroimaging is equally as important in patients who suffer an SAH. DSA remains the reference standard for imaging of intracranial aneurysms due to its high spatial and temporal resolution. As noninvasive imaging technology, such as CTA and MRA, improves, the diagnostic accuracy of such tests continues to increasingly approximate that of DSA. In cases of angiographically negative SAH, imaging protocols are necessary not only for diagnosis but also to search for an initially occult vascular lesion, such as a thrombosed, ruptured aneurysm that might be detected in a delayed fashion. Given the crucial role of neuroimaging in all aspects of care for patients with UIAs and SAH, it is incumbent on those who care for these patients, including cerebrovascular neurosurgeons, interventional neurologists and neuroradiologists, and diagnostic radiologists and neurointensivists, to understand the role of imaging in this disease and how individual members of the multispecialty team use imaging to ensure best practices to deliver cutting-edge care to these often complex cases. This review expounds on the role of imaging in the management of UIAs and ruptured intracranial aneurysms and in the workup of angiographically negative subarachnoid hemorrhage.

Intra- and inter-observer variability in intracranial aneurysm segmentation: comparison between CT angiography (semi-automated segmentation software stroke VCAR) and digital subtraction angiography (3D rotational angiography)

PURPOSE

To compare size and morphologic features of three-dimensional aneurysm models, obtained with a semi-automated segmentation software (Stroke VCAR, GE, USA) from cerebral CT angiography (CTA) data, to three-dimensional aneurysm models obtained with digital subtraction angiography (DSA, with 3D rotational angiography acquisition-3DRA), considered as the reference standard.

METHODS

In this retrospective study, we reviewed 132 patients, with a total number of 137 intracranial aneurysm, who underwent CTA and subsequent DSA examination, supplemented with 3DRA. We compared neck length, short axis and long axis measured on 3DRA model to the same variables measured on 3D-CTA model by two blinded readers and to the automatic software dimensions. Therefore, statistics analysis assessed intra-observer and inter-observer variability and differences between patients with or without subarachnoid hemorrhage (SAH).

RESULTS

There were no significant differences in short-axis and long-axis measurements between 3D angiographic and 3D-CTA models, while comparison of neck lengths revealed a statistically significant difference, which tended to be greater for smaller neck lengths (partial volume effect and "kissing vessels" artifact). There were significant differences between manual and automatic data measured for the same three variables, and the presence of SAH did not affect aneurysm 3D reconstruction. Inter-observer agreement resulted moderate for neck length and substantial for short axis and long axis.

CONCLUSION

The examined 3D-CTA segmentation system is a reproducible procedure for aneurysm morphologic characterization and, in particular, for assessment of aneurysm sac dimensions, but considerable carefulness is required in neck length interpretation.

Features of "false positive" unruptured intracranial aneurysms on screening magnetic resonance angiography

BACKGROUND

Physicians can find it challenging to decide whether confirmative digital subtraction angiography (DSA) should be performed in patients who present with "suspicious small aneurysm-like structures" on magnetic resonance angiography (MRA). Factors associated with "false positive aneurysms on MRA" (FPAMs)," which are finally confirmed as negative on DSA, have rarely been reported. This study aimed to identify the clinical or radiologic clues indicative of FPAM on DSA.

METHODS

Patients who had undergone DSA between 2016 and 2019 for suspicious aneurysm-like structures < 5 mm in size on MRA were enrolled. Patient demographics and the details regarding the geometry of the structures were retrospectively reviewed. Univariate and multivariate logistic regression analyses were conducted to identify the associated factors. Receiver operating characteristic curve analysis was performed to assess the clinical implications.

RESULTS

Of the 107 suspicious structures, 46 were indicated as being false positive on DSA (42.96%). Location (positive on C7 and negative on C5-6 ICA) and lower dome to neck ratio were found to be significant parameters in the multivariate analysis. The dome to neck ratio threshold value was 0.99.

CONCLUSION

Suspicious aneurysm-like structures located not on C5-6 but on C7 ICA and having wide neck morphologies (dome to neck ratio < 0.99) are highly likely to be negative on DSA.

Hemodynamics of Cerebral Aneurysms: Connecting Medical Imaging and Biomechanical Analysis

In the last two decades, numerous studies have conducted patient-specific computations of blood flow dynamics in cerebral aneurysms and reported correlations between various hemodynamic metrics and aneurysmal disease progression or treatment outcomes. Nevertheless, intra-aneurysmal flow analysis has not been adopted in current clinical practice, and hemodynamic factors usually are not considered in clinical decision making. This review presents the state of the art in cerebral aneurysm imaging and image-based modeling, discussing the advantages and limitations of each approach and focusing on the translational value of hemodynamic analysis. Combining imaging and modeling data obtained from different flow modalities can improve the accuracy and fidelity of resulting velocity fields and flow-derived factors that are thought to affect aneurysmal disease progression. It is expected that predictive models utilizing hemodynamic factors in combination with patient medical history and morphological data will outperform current risk scores and treatment guidelines. Possible future directions include novel approaches enabling data assimilation and multimodality analysis of cerebral aneurysm hemodynamics.

Multiple ruptured cerebral aneurysms at the National Hospital of the Kyrgyz Republic between 2008 and 2014: a departmental summary

Despite the rapid progress of vascular neurosurgery with the development of microsurgical and endovascular techniques, the optimal strategy for surgical treatment of multiple cerebral aneurysms has not yet been developed. The indications for choosing one-stage or multi-stage surgery remain unsolved. This is a summary of the departmental routine reports at the Clinic of Neurosurgery, National Hospital of the Kyrgyz Republic. Subjects were 235 patients (124 males and 111 females) with ruptured multiple cerebral aneurysms admitted to the hospital. Their ages ranged from 18 to 72 years (average and standard deviation: 44.3 ± 9.7 years) and 48.1% of patients had 3 or more aneurysms. Among aneurysms that ruptured, 20.4% were a giant aneurysm (>25 mm) and 43.0% of patients had grade IV or V according to the Hunt-Hess Scale. Among 228 patients who were operated on, 147 were treated by single-stage surgery and 81 by multi-stage surgery. Microsurgical operations with clipping of the aneurysm neck were performed in 141 (61.8%) patients (97 single-stage and 44 multi-stage), while 40 (17.5%) patients (16 single-stage and 24 multi-stage) were operated using the endovascular technique. The number of palliative surgeries (trapping, ligation of the internal carotid artery, and reinforcement of the aneurysm wall) was significantly less (p=0.011) with multi-stage surgery (9 out of 81 cases, 11.1%) than with single-stage surgery (38 out of 147 cases, 25.9%). Among 600 aneurysms, 583 (97.2%) were treated by either single-stage surgery (n=296) or multi-stage surgery (n=287). There were no differences in prognosis at discharge between single-stage and multi-stage surgery.

Deep learning for automated cerebral aneurysm detection on computed tomography images

PURPOSE

Cerebrovascular aneurysms are being observed with rapidly increasing incidence. Therefore, tools are needed for accurate and efficient detection of aneurysms. We used deep learning techniques with CT angiography acquired from multiple medical centers and different machines to develop and evaluate an automatic detection model.

METHODS

In this study, we have introduced a deep learning model, the faster RCNN model, in order to develop a tool for automatic detection of aneurysms from medical images. The inputs of the model were 2D nearby projection (NP) images from 3D CTA, which were made by the NP method proposed in this study. This method made aneurysms clearly visible on images and improved the model's performance. The study included 311 patients with 352 aneurysms, selected from three hospitals, and 208 and 103 of these patients, respectively, were randomly selected to train and test the models.

RESULTS

The sensitivity of the trained model was 91.8%. For aneurysm sizes larger than 3 mm, the sensitivity of successful aneurysm detection was 96.7%. We achieved state-of-the-art sensitivity for > 3 mm aneurysms. The sensitivities also indicated that there was no significant difference among aneurysms at different locations in the body. Computing time for the detection process was less than 25 s per case.

CONCLUSIONS

We successfully developed a deep learning model that can automatically detect aneurysms. The model performed well for aneurysms of different sizes or in different locations. This finding indicates that the deep learning model has the potential to vastly improve clinician performance by providing automated aneurysm detection.

Subarachnoid haemorrhage with negative initial neurovascular imaging: a systematic review and meta-analysis

Background

In patients with spontaneous subarachnoid haemorrhage (SAH), a vascular cause for the bleed is not always found on initial investigations. This study aimed to systematically evaluate the delayed investigation strategies and clinical outcomes in these cases, often described as “non-aneurysmal” SAH (naSAH).

Methods

A systematic review was performed in concordance with the PRISMA checklist. Pooled proportions of primary outcome measures were estimated using a random-effects model.

Results

Fifty-eight studies were included (4473 patients). The cohort was split into perimesencephalic naSAH (PnaSAH) (49.9%), non-PnaSAH (44.7%) and radiologically negative SAH identified on lumbar puncture (5.4%). The commonest initial vascular imaging modality was digital subtraction angiography. A vascular abnormality was identified during delayed investigation in 3.9% [95% CI 1.9–6.6]. There was no uniform strategy for the timing or modality of delayed investigations. The pooled proportion of a favourable modified Rankin scale outcome (0–2) at 3–6 months following diagnosis was 92.0% [95% CI 86.0–96.5]. Complications included re-bleeding (3.1% [95% CI 1.5–5.2]), hydrocephalus (16.0% [95% CI 11.2–21.4]), vasospasm (9.6% [95% CI 6.5–13.3]) and seizure (3.5% [95% CI 1.7–5.8]). Stratified by bleeding pattern, we demonstrate a higher rate of delayed diagnoses (13.6% [95% CI 7.4–21.3]), lower proportion of favourable functional outcome (87.2% [95% CI 80.1–92.9]) and higher risk of complications for non-PnaSAH patients.

Conclusion

This study highlights the heterogeneity in delayed investigations and outcomes for patients with naSAH, which may be influenced by the initial pattern of bleeding. Further multi-centre prospective studies are required to clarify optimal tailored management strategies for this heterogeneous group of patients.

Electronic supplementary material

The online version of this article (10.1007/s00701-019-04025-w) contains supplementary material, which is available to authorized users.

Delayed diagnosis of post-traumatic aneurysm of distal anterior cerebral artery

BACKGROUND

Traumatic intracranial aneurysms (TICA) are often associated with poor prognosis and should be diagnosed as soon as possible to prevent delayed intracranial hemorrhage and high rates of morbidity/mortality related to bleeding. Diagnosis requires a high index of suspicion. The goal of treatment is to exclude the aneurysm issue with surgical or endovascular methods.

CASE DESCRIPTION

We report the case of a 19-year-old boy who suffered a cranio-orbital trauma; 2 weeks after initial trauma he deteriorates with a new intracranial bleeding. Immediate angiography resulted negative. Delayed follow-up by magnetic resonance angiography showed an unruptured aneurysm of anterior cerebral artery that was successfully clipped.

CONCLUSIONS

A TICA should be suspected in case of delayed deterioration in head-injured patient, prompt diagnosis and treatment could improve prognosis and reduce morbidity and mortality.

Systematic Review of the Literature: Best Practices

Reviews of published scientific literature are a valuable resource that can underline best practices in medicine and clarify clinical controversies. Among the various types of reviews, the systematic review of the literature is ranked as the most rigorous since it is a high-level summary of existing evidence focused on answering a precise question. Systematic reviews employ a pre-defined protocol to identify relevant and trustworthy literature. Such reviews can accomplish several critical goals that are not easily achievable with typical empirical studies by allowing identification and discussion of best evidence, contradictory findings, and gaps in the literature. The Association of University Radiologists Radiology Research Alliance Systematic Review Task Force convened to explore the methodology and practical considerations involved in performing a systematic review. This article provides a detailed and practical guide for performing a systematic review and discusses its applications in radiology.

Digital Subtraction Angiography for CT Angiogram Negative Haemorrhages

BACKGROUND

In patients with subarachnoid haemorrhage (SAH) and a negative finding on CT angiography (CTA), further imaging with digital subtraction angiography (DSA) is commonly performed to identify the source of bleeding. The purpose of this study was to investigate whether negative findings on CTA can reliably exclude aneurysms in patients with acute SAH.

METHODS

This retrospective study identified all DSAs performed between August 2010 and July 2014 within our institution. CT angiography was performed with a 64-section multidetector row CT scanner. Only DSAs from patients with confirmed SAH and a negative CTA result were included in the final analyses. A fellowship-trained neuroradiologist reviewed the imaging results.

RESULTS

Of the 857 DSAs, 50 (5.83%) were performed in 35 patients with CTA-negative SAH. Of the 35 patients, three (8.57%) had positive findings on the DSA. In one patient, suspicious dissection of the extra- and intra-cranial segment of the right vertebral artery could not be confirmed even in retrospect. In the second patient, the suspicious finding of tiny protuberance from the left paraclinoid internal carotid artery (ICA) on DSA did not change on follow-up and did not change patient's management. The third patient had a posterior inferior cerebellar artery aneurysm, which was not seen on the initial CTA owing to the incomplete coverage of the head on the CTA.

CONCLUSION

In patients with SAH, negative findings on a technically sound CTA are reliable in ruling out aneurysms in any pattern of SAH or no blood on CT. Our observations need to be confirmed with larger prospective studies.

ACR Appropriateness Criteria® Cerebrovascular Disease

Diseases of the cerebral vasculature represent a heterogeneous group of ischemic and hemorrhagic etiologies, which often manifest clinically as an acute neurologic deficit also known as stroke or less commonly with symptoms such as headache or seizures. Stroke is the fourth leading cause of death and is a leading cause of serious long-term disability in the United States. Eighty-seven percent of strokes are ischemic, 10% are due to intracerebral hemorrhage, and 3% are secondary to subarachnoid hemorrhage. The past two decades have seen significant developments in the screening, diagnosis, and treatment of ischemic and hemorrhagic causes of stroke with advancements in CT and MRI technology and novel treatment devices and techniques. Multiple different imaging modalities can be used in the evaluation of cerebrovascular disease. The different imaging modalities all have their own niches and their own advantages and disadvantages in the evaluation of cerebrovascular disease. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

Accuracy of detecting enlargement of aneurysms using different MRI modalities and measurement protocols

OBJECTIVE

Aneurysm growth is considered predictive of future rupture of intracranial aneurysms. However, how accurately neuroradiologists can reliably detect incremental aneurysm growth using clinical MRI is still unknown. The purpose of this study was to assess the agreement rate of detecting aneurysm enlargement employing generally used MRI modalities.

METHODS

Three silicone flow phantom models, each with 8 aneurysms of various sizes at different sites, were used in this study. The aneurysm models were identical except for an incremental increase in the sizes of the 8 aneurysms, which ranged from 0.4 mm to 2 mm. The phantoms were imaged on 1.5-T and 3-T MRI units with both time-of-flight (TOF) and contrast-enhanced MR angiography. Three independent expert neuroradiologists measured the aneurysms in a blinded manner using different measurement approaches. The individual and agreement detection rates of aneurysm enlargement among the 3 experts were calculated.

RESULTS

The mean detection rate of any increase in any aneurysmal dimension was 95.7%. The detection rates of the 3 observers (observers A, B, and C) were 98.0%, 96.6%, and 92.7%, respectively (p = 0.22). The detection rates of each MRI modality were 91.3% using 1.5-T TOF, 97.2% using 1.5-T with Gd, 95.8% using 3.0-T TOF, and 97.2% using 3.0-T with Gd (p = 0.31). On the other hand, the mean detection rate for aneurysm enlargement was 54.8%. Specifically, the detection rates of observers A, B, and C were 49.0%, 46.1%, and 66.7%, respectively (p = 0.009). As the incremental enlargement value increased, the detection rate for aneurysm enlargement increased. The use of 1.5-T Gd improved the detection rate for small incremental enlargement (e.g., 0.4–1 mm) of the aneurysm (p = 0.04). The location of the aneurysm also affected the detection rate for aneurysm enlargement (p < 0.0001).

CONCLUSIONS

The detection rate and interobserver agreement were very high for aneurysm enlargement of 0.4–2 mm. The detection rate for at least 1 increase in any aneurysm dimension did not depend on the choice of MRI modality or measurement protocol. Use of Gd improved the accuracy of measurement. Aneurysm location may influence the accuracy of detecting enlargement.