ABSTRACT
BACKGROUND: Limited research exists regarding the impact of neuroimaging on endovascular thrombectomy (EVT) decisions for late-window cases of large vessel occlusion (LVO) stroke. OBJECTIVE: T0 assess whether perfusion CT imaging: (1) alters the proportion of recommendations for EVT, and (2) enhances the reliability of EVT decision-making compared with non-contrast CT and CT angiography. METHODS: We conducted a survey using 30 patients drawn from an institutional database of 3144 acute stroke cases. These were presented to 29 Canadian physicians with and without perfusion imaging. We used non-overlapping 95% confidence intervals and difference in agreement classification as criteria to suggest a difference between the Gwet AC1 statistics (κG). RESULTS: The percentage of EVT recommendations differed by 1.1% with or without perfusion imaging. Individual decisions changed in 21.4% of cases (11.3% against EVT and 10.1% in favor). Inter-rater agreement (κG) among the 29 raters was similar between non-perfusion and perfusion CT neuroimaging (κG=0.487; 95% CI 0.327 to 0.647 and κG=0.552; 95% CI 0.430 to 0.675). The 95% CIs overlapped with moderate agreement in both. Intra-rater agreement exhibited overlapping 95% CIs for all 28 raters. κG was either substantial or excellent (0.81-1) for 71.4% (20/28) of raters in both groups. CONCLUSIONS: Despite the minimal difference in overall EVT recommendations with either neuroimaging protocol one in five decisions changed with perfusion imaging. Regarding agreement we found that the use of automated CT perfusion images does not significantly impact the reliability of EVT decisions for patients with late-window LVO.
ABSTRACT
This study conducts collateral evaluation from ischemic damage using a deep learning-based Siamese network, addressing the challenges associated with a small and imbalanced dataset. The collateral network provides an alternative oxygen and nutrient supply pathway in ischemic stroke cases, influencing treatment decisions. Research in this area focuses on automated collateral assessment using deep learning (DL) methods to expedite decision-making processes and enhance accuracy. Our study employed a 3D ResNet-based Siamese network, referred to as SCANED, to classify collaterals as good/intermediate or poor. Utilizing non-contrast computed tomography (NCCT) images, the network automates collateral identification and assessment by analyzing tissue degeneration around the ischemic site. Relevant features from the left/right hemispheres were extracted, and Euclidean Distance (ED) was employed for similarity measurement. Finally, dichotomized classification of good/intermediate or poor collateral is performed by SCANED using an optimal threshold derived from ROC analysis. SCANED provides a sensitivity of 0.88, a specificity of 0.63, and a weighted F1 score of 0.86 in the dichotomized classification.
Subject(s)
Brain Ischemia , Ischemic Stroke , ROC Curve , Brain Ischemia/diagnosis , Deep Learning , Ischemic Stroke/diagnosis , HumansABSTRACT
BACKGROUND: A significant portion of cryptogenic stroke is hypothesized to be secondary to cardiac embolism. However, transthoracic echocardiogram is usually delayed after stroke, and more detailed cardiac imaging is not routinely done. AIMS: This study aimed to determine whether non-ECG-gated cardiac CT angiography (cCTA) during hyperacute stroke would provide diagnostic quality images and act as an adjunct modality of cardiac imaging to detect sources of emboli. METHODS: In this single-center prospective cohort study, modified Code Stroke imaging was implemented with a 64-slice CT scanner, where the longitudinal axis of CT angiography was extended from the carina to the diaphragm. The primary outcomes of image quality, recruitment feasibility, impact on hyperacute time metrics, and additional radiation dose were assessed. Secondary outcomes consisted of detection of high-risk cardiac sources of embolism, mediastinal or lung pathology, and impact on etiologic classification. RESULTS: One hundred and twenty eligible patients were enrolled, of which 105 (87.5%) had good/moderate quality images for motion artifact and 119 (99.2%) for contrast opacification. Total CT time, door-to-needle time, and door-to-groin puncture time were unchanged with the addition of cCTA. Eighty-nine patients received a final diagnosis of ischemic stroke, of which 12/89 (13.5%) had high-risk cardioembolic findings on cCTA. Incidental findings, such as pulmonary embolism (PE) (7/89, 7.9%) and malignancy (6/89, 6.7%), were observed. cCTA led to changes in management for 19/120 (15.8%) of all patients, and reclassification of stroke etiology for 8/89 (9%) of patients. CONCLUSIONS: Non-ECG-gated cCTA can be feasibly incorporated into Code Stroke and provide diagnostic quality images without delays in hyperacute time metrics. It can detect high-risk cardiac sources, and other findings impacting patient care. This may help reclassify a subset of cryptogenic stroke cases and improve secondary prevention.
Subject(s)
Embolism , Ischemic Stroke , Stroke , Humans , Stroke/diagnostic imaging , Stroke/etiology , Computed Tomography Angiography/methods , Prospective Studies , Embolism/complications , Ischemic Stroke/complications , Coronary Angiography/adverse effects , Coronary Angiography/methods , Radiation DosageABSTRACT
OBJECTIVE: To conduct feasibility and cost analysis of portable MRI implementation in a remote setting where MRI access is otherwise unavailable. METHODS: Portable MRI (ultra-low field, 0.064T) was installed in Weeneebayko General Hospital, Moose Factory, Ontario. Adult patients, presenting with any indication for neuroimaging, were eligible for study inclusion. Scanning period was from November 14, 2021, to September 6, 2022. Images were sent via a secure PACS network for Neuroradiologist interpretation, available 24/7. Clinical indications, image quality, and report turnaround time were recorded. A cost analysis was conducted from a healthcare system's perspective in 2022 Canadian dollars, comparing cost of portable MRI implementation to transporting patients to a center with fixed MRI. RESULTS: Portable MRI was successfully implemented in a remote Canadian location. Twenty-five patients received a portable MRI scan. All studies were of diagnostic quality. No clinically significant pathologies were identified on any of the studies. However, based on clinical presentation and limitations of portable MRI resolution, it is estimated that 11 (44%) of patients would require transfer to a center with fixed MRI for further imaging workup. Cost savings were $854,841 based on 50 patients receiving portable MRI over 1 year. Five-year budget impact analysis showed nearly $8 million dollars saved. CONCLUSIONS: Portable MRI implementation in a remote setting is feasible, with significant cost savings compared to fixed MRI. This study may serve as a model to democratize MRI access, offer timely care and improved triaging in remote areas where conventional MRI is unavailable.
ABSTRACT
PURPOSE: This study aims to increase awareness of the hemorrhagic presentation of intracranial meningiomas in the emergency department and present clues for neuroradiological diagnosis, which is crucial for pertinent management. We described the prevalence of hemorrhage in a large population of meningioma patients, with emphasis on clinical presentation, computed tomography (CT), magnetic resonance (MR), and digital subtraction angiography (DSA) findings. METHODS: This retrospective analysis has been performed at two reference institutions between January 2002 and December 2015, and includes 1304 patients with histologically proven newly diagnosed intracranial meningioma. Clinical features and neuroradiological findings of intracranial meningiomas presenting with hemorrhage have been reviewed. RESULTS: Twenty-four patients (1.8%, 16 females, 8 males, age range: 29-88 years) were found to have spontaneous hemorrhagic onset of the newly diagnosed meningioma. A sudden onset occurred in 23/24 patients. Sixteen patients showed isolated intralesional hemorrhage, four had subdural hematomas, and the remaining four presented combined intralesional and subarachnoid (n = 2) or intraventricular (n = 2) hemorrhages. In 13 patients, CT showed both the hemorrhage and the meningioma. In the other 11 patients, diagnosis was achieved by emergency or early surgery (n = 5), MRI (n = 5), and DSA (n = 1). CONCLUSIONS: The presence of an underlying meningioma has to be considered in the differential diagnosis of spontaneous intracranial hemorrhage, although this is a rare event. CT, MRI, and occasionally DSA were useful to obtain the diagnosis; however, in up to a fifth of patients, this was achieved at surgery.
Subject(s)
Meningeal Neoplasms , Meningioma , Male , Female , Humans , Adult , Middle Aged , Aged , Aged, 80 and over , Meningioma/diagnostic imaging , Meningioma/surgery , Retrospective Studies , Tomography, X-Ray Computed/methods , Hematoma, Subdural , Meningeal Neoplasms/diagnostic imaging , Meningeal Neoplasms/pathology , Meningeal Neoplasms/surgery , Angiography, Digital Subtraction/methodsABSTRACT
PURPOSE: Collateral evaluation is typically done using visual inspection of cerebral images and thus suffers from intra- and inter-rater variability. Large open databases of ischemic stroke patients are rare, limiting the use of deep learning methods in treatment decision-making. METHODS: We adapted a pre-trained EfficientNet B0 network through transfer learning to improve collateral evaluation using slice-based and subject-level classification. Our method uses stacking and overlapping of 2D slices from a patient's 4D computed tomography angiography (CTA) and a majority voting scheme to determine a patient's final collateral grade based on all classified 2D MIPs. Class imbalance is handled in the evaluation process by using the focal loss with class weight to penalize the majority class. RESULTS: We evaluated our method using a nine-fold cross-validation performed with 83 subjects. Mean sensitivity of 0.71, specificity of 0.84, and a weighted F1 score of 0.71 in multi-class (good, intermediate, and poor) classification were obtained. Considering treatment effect, a dichotomized decision is also made for collateral scoring of a subject based on two classes (good/intermediate and poor) which achieves a sensitivity of 0.89 and specificity of 0.96 with a weighted F1 score of 0.95. CONCLUSION: An automatic and robust collateral assessment method that mitigates the issues with the small imbalanced dataset was developed. Computer-aided evaluation of collaterals can help decision-making of ischemic stroke treatment strategy in clinical settings.
Subject(s)
Brain Ischemia , Deep Learning , Ischemic Stroke , Stroke , Humans , Stroke/diagnostic imaging , Stroke/therapy , Cerebral Angiography/methods , Computed Tomography Angiography/methods , Four-Dimensional Computed Tomography/methods , Brain Ischemia/diagnostic imaging , Brain Ischemia/therapy , Retrospective StudiesABSTRACT
OBJECTIVE: The objective of this scoping review is to review the available information on objective performance metrics used during robotic neuroendovascular intervention procedures on humans. INTRODUCTION: Robotic neuroendovascular intervention is defined as any endovascular procedure within the vasculature of the central nervous system with the assistance of a robotic system for diagnostic or therapeutic procedures. Robotic systems are described as a 2-component system consisting of a patient-side mechanical robot, and a separate operator control station. Robotic neuroendovascular intervention is a growing field and there is a need to establish objective performance metrics for furthering evidence-based reporting of the literature. INCLUSION CRITERIA: This scoping review will consider all studies involving humans that utilize robotic neuroendovascular intervention. We will consider all types of studies, reports, and reviews as well as gray literature. Studies will be included if they describe the use of an objective performance metric during robotic neuroendovascular intervention. This review is not limited to a particular country or health care system, and will consider all study designs, regardless of their rigor or language. METHODS: Utilizing a 3-step framework as a guide, we will perform a systematic search in Embase, Cochrane Library, and MEDLINE. Available literature from inception to the present will be considered. Studies will be independently screened according to the inclusion criteria by 2 reviewers based on title, abstract, and full text. Data will be extracted, sorted, and presented in both a narrative summary as well as table and diagram based on the objective of the scoping review.
Subject(s)
Endovascular Procedures , Robotic Surgical Procedures , Robotics , Humans , Research Design , Robotic Surgical Procedures/methods , Systematic Reviews as TopicABSTRACT
Introduction: Adult genetic leukoencephalopathies are rare neurological disorders that present unique diagnostic challenges due to their clinical and radiological overlap with more common white matter diseases, notably multiple sclerosis (MS). In this context, a strong collaborative multidisciplinary network is beneficial for shortening the diagnostic odyssey of these patients and preventing misdiagnosis. The White Matter Rounds (WM Rounds) are multidisciplinary international online meetings attended by more than 30 physicians and scientists from 15 participating sites that gather every month to discuss patients with atypical white matter disorders. We aim to present the experience of the WM Rounds Network and demonstrate the value of collaborative multidisciplinary international case discussion meetings in differentiating and preventing misdiagnoses between genetic white matter diseases and atypical MS. Methods: We retrospectively reviewed the demographic, clinical and radiological data of all the subjects presented at the WM Rounds since their creation in 2013. Results: Seventy-four patients (mean age 44.3) have been referred and discussed at the WM Rounds since 2013. Twenty-five (33.8%) of these patients were referred by an MS specialist for having an atypical presentation of MS, while in most of the remaining cases, the referring physician was a geneticist (23; 31.1%). Based on the WM Rounds recommendations, a definite diagnosis was made in 36/69 (52.2%) patients for which information was available for retrospective review. Of these diagnosed patients, 20 (55.6%) had a genetic disease, 8 (22.2%) had MS, 3 (8.3%) had both MS and a genetic disorder and 5 (13.9%) had other non-genetic conditions. Interestingly, among the patients initially referred by an MS specialist, 7/25 were definitively diagnosed with MS, 5/25 had a genetic condition (e.g., X-linked adrenoleukodystrophy and hereditary small vessel diseases like Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy (CADASIL) and COL4A1-related disorder), and one had both MS and a genetic demyelinating neuropathy. Thanks to the WM Rounds collaborative efforts, the subjects who currently remain without a definite diagnosis, despite extensive investigations performed in the clinical setting, have been recruited in research studies aimed at identifying novel forms of genetic MS mimickers. Conclusions: The experience of the WM Rounds Network demonstrates the benefit of collective discussions on complex cases to increase the diagnostic rate and decrease misdiagnosis in patients with rare or atypical white matter diseases. Networks of this nature allow physicians and scientists to compare and share information on challenging cases from across the world, provide a basis for future multicenter research studies, and serve as model for other rare diseases.
ABSTRACT
Epidermoid cysts are benign congenital extra-axial lesions commonly found in the posterior fossa. These lesions have a characteristic imaging appearance on computed tomography (CT) scan and magnetic resonance imaging (MRI), but occasionally they may exhibit atypical radiological features, showing unusual hyperintensity on T1-weighted images (T1WI). Currently, such atypical appearance is referred to as white epidermoid. We present the imaging features of 5 cases of white epidermoid cyst and discuss the possible underlying etiology of this unusual radiological appearance. We retrospectively searched our electronic radiology database from January 2005 to December 2015 for all intracranial epidermoid cysts, which were confirmed either by typical MRI appearance or histopathological examination. All white epidermoid cases were evaluated with non-enhanced CT scan and multisequential MRI. Histopathological correlation was carried out in four white epidermoid cases. A total of 61 patients with epidermoid cyst were found, of those 5 (8%) were considered white epidermoids. These consisted of 3 females and 2 males, ranging in age between 31-63 years (average age was 51.8 years). Three patients had lesions located in the posterior fossa. The 2 other patients had lesions in the suprasellar region, with extension to the right middle cranial fossa in one. All 5 lesions were hyperdense on CT scan and hyperintense on T1WI. One patient demonstrated evidence of transformation of a classic epidermoid to a white epidermoid after partial resection. Histopathologically, cholesterol clefts were seen in 3 epidermoid cysts, each which also showed microcalcifications, proteinaceous material or melanin. Hemorrhage was demonstrated in one additional lesion. White epidermoid cyst is an unusual intracranial lesion that should be considered when encountered with an extra-axial T1 hyperintense lesion. The cause of this hyperintensity is not clearly understood, but the presence of cholesterol, microcalcifications, proteinaceous content and rarely hemorrhage or melanin may be contributing factors.
Subject(s)
Brain Diseases/diagnostic imaging , Brain Diseases/pathology , Epidermal Cyst/diagnostic imaging , Epidermal Cyst/pathology , Adult , Cranial Fossa, Posterior , Diffusion Tensor Imaging , Female , Humans , Magnetic Resonance Imaging , Male , Middle Aged , Tomography, X-Ray ComputedABSTRACT
A previously healthy 48-year-old female presented to the emergency department with a 2-week history of low back pain, progressive lower extremities weakness, and right leg numbness. There were no bowel or bladder dysfunction symptoms. Spine magnetic resonance imaging (MRI) showed an intradural cystic lesion dorsal to the spinal cord at the level of L1 measuring 1.6 × 2.1 × 4.1 cm, which was T1 hypointense and T2 hyperintense, with a small soft tissue component and no gadolinium enhancement (Figure 1). A small lipomatous component was also noted. There were no associated vertebral anomalies. The patient underwent a T12-L2 laminectomy and cyst resection, which was subtotal due to the cyst adherence to the conus medullaris. Histopathology showed characteristic features of a neurenteric cyst, with respiratory-type epithelium in the cyst wall (Figure 2). Eight months later, follow-up MRI showed no evidence of recurrence. The patient reported improved sensation in the lower extremities; however, there was some residual weakness predominantly in the proximal hip flexors bilaterally.
Subject(s)
Neural Tube Defects , Female , Gadolinium , Humans , Lumbosacral Region , Magnetic Resonance Imaging , Middle Aged , Neural Tube Defects/diagnostic imaging , Neural Tube Defects/surgery , Spinal CordABSTRACT
BACKGROUND: No randomized trial of intracranial aneurysm coiling has compared long-term efficacy of polymer-modified coils to bare metal coils (BMCs). We report 5-year results comparing Matrix2 coils to BMCs. The primary objective was to compare the rates of target aneurysm recurrence (TAR) at 12 months. Secondary objectives included angiographic outcomes at TAR or 12 months and TAR at 5 years. METHODS: A total of 626 patients were randomized to BMCs or Matrix2 coils. Detailed methods and 1-year results have been published previously. RESULTS: Of 580 patients eligible for 5-year follow-up, 431 (74.3%) completed follow-up or reached TAR. Matrix2 coils were non-inferior to BMCs (P=0.8) but did not confer any benefit. Core lab reported post-treatment residual aneurysm filling (Raymond III) correlated with TAR (P<0.0001) and with aneurysm hemorrhage after treatment (P<0.008). Repeat aneurysmal hemorrhage after treatment, but before hospital discharge, occurred in three patients treated for acutely ruptured aneurysms. Additionally, two patients treated for unruptured aneurysms experienced a first hemorrhage during follow-up. All five hemorrhages resulted from aneurysms with Raymond III residual aneurysm filling persisting after initial treatment. After 5 years follow-up, 2/626 (0.3%) patients are known to have had target aneurysm rupture following hospital discharge. The annualized rate of delayed hemorrhage after coiling was 2/398/5=0.001 (0.1%) per year for unruptured aneurysms and 0 for ruptured aneurysms. CONCLUSIONS: After 5 years Matrix2 coils were non-inferior to BMCs but no benefit was demonstrated. Post-treatment residual angiographic aneurysm filling (Raymond III) is strongly associated with TAR (P<0.0001) and post-treatment aneurysmal hemorrhage (P=0.008).
Subject(s)
Aneurysm, Ruptured , Embolization, Therapeutic , Intracranial Aneurysm , Aneurysm, Ruptured/therapy , Embolization, Therapeutic/adverse effects , Humans , Intracranial Aneurysm/diagnostic imaging , Intracranial Aneurysm/therapy , Treatment OutcomeABSTRACT
Cerebral peri-aneurysmal edema (PE) is typically associated with giant partially-thrombosed aneurysms and less frequently with smaller aneurysms treated with endovascular embolization. An understanding of the pathophysiologic mechanism of PE is still limited. We report 3 cases of cerebral aneurysms associated with PE. We describe 2 cases of giant partially thrombosed aneurysms surrounded by vasogenic edema with apposition of an intramural and juxtamural thrombus. Our third case is a smaller aneurysm inciting vasogenic edema several years after coil embolization. Vessel-wall magnetic resonance imaging (MRI) showed avid wall enhancement and an enhancing thrombus embedded within the coils, reflecting inflammation of the aneurysm wall and proliferation of the vasa vasorum. Thrombosis within the aneurysmal sac and walls, both in native and treated aneurysms, may promote inflammatory changes and sustain the occurrence of PE. Vessel-wall MRI has a potential role in the evaluation process of this subgroup of aneurysms.
ABSTRACT
PURPOSE: Sufficient collateral blood supply is crucial for favorable outcomes with endovascular treatment. The current practice of collateral scoring relies on visual inspection and thus can suffer from inter and intra-rater inconsistency. We present a robust and automatic method to score cerebral collateral blood supply to aid ischemic stroke treatment decision making. The developed method is based on 4D dynamic CT angiography (CTA) and the ASPECTS scoring protocol. METHODS: The proposed method, ACCESS (Automatic Collateral Circulation Evaluation in iSchemic Stroke), estimates a target patient's unfilled cerebrovasculature in contrast-enhanced CTA using the lack of contrast agent due to clotting. To do so, the fast robust matrix completion algorithm with in-face extended Frank-Wolfe optimization is applied on a cohort of healthy subjects and a target patient, to model the patient's unfilled vessels and the estimated full vasculature as sparse and low-rank components, respectively. The collateral score is computed as the ratio of the unfilled vessels to the full vasculature, mimicking existing clinical protocols. RESULTS: ACCESS was tested with 46 stroke patients and obtained an overall accuracy of 84.78%. The optimal threshold selection was evaluated using a receiver operating characteristics curve with the leave-one-out approach, and a mean area under the curve of 85.39% was obtained. CONCLUSION: ACCESS automates collateral scoring to mitigate the shortcomings of the standard clinical practice. It is a robust approach, which resembles how radiologists score clinical scans, and can be used to help radiologists in clinical decisions of stroke treatment.
Subject(s)
Brain Ischemia/diagnostic imaging , Cerebral Angiography , Collateral Circulation , Computed Tomography Angiography , Diagnosis, Computer-Assisted/methods , Four-Dimensional Computed Tomography , Ischemic Stroke/diagnostic imaging , Aged , Algorithms , Decision Making , Female , Healthy Volunteers , Humans , Image Processing, Computer-Assisted/methods , Machine Learning , Male , Middle Aged , ROC CurveABSTRACT
PURPOSE: Developmental in nature, brain arteriovenous malformations (AVM) have the potential to affect whole brain organization. Here we investigated the impact of AVM on functional and structural brain organization using resting-state functional MRI (rsfMRI) and cortical thickness measures. METHODS: We investigated brain functional organization and structure using rsfMRI in conjunction with cortical thickness analyses in 23 patients with cerebral arteriovenous malformations (AVMs) and 20 healthy control subjects. RESULTS: Healthy controls showed the expected anti-correlation between activity in the default mode network (DMN) and frontal areas that are part of the attentional control network. By contrast, patients demonstrated a disruption of this anti-correlation. Disruptions to this anti-correlation were even observed in a subgroup of patients with lesions remote from the main nodes of the DMN and were unrelated to differences in perfusion. Functional connectivity differences were accompanied by reduced cortical thickness in frontal attentional areas in patients compared to the controls. CONCLUSIONS: These results contribute to the discussion that AVMs affect whole brain networks and not simply the area surrounding the lesion.