Clinical significance of intracranial hemorrhage after thrombectomy detected solely by magnetic resonance imaging and not by computed tomography.

BACKGROUND AND OBJECTIVE: Whether intracranial hemorrhage (ICH) detected using magnetic resonance imaging (MRI) affects the clinical outcomes of patients with large-vessel occlusion (LVO) treated with mechanical thrombectomy (MT) remains unclear. This study investigated the clinical features of ICH after MT detected solely by MRI. METHODS: This was a retrospective analysis of patients with acute ischemic stroke and occlusion of the internal carotid artery or middle cerebral artery treated with MT between April 2011 and March 2021. Among 632 patients, patients diagnosed with no ICH using CT, with a pre-morbid modified Rankin Scale (mRS) score ≤ 2, and those who underwent MRI including T2* and computed tomography (CT) within 72 h from MT were enrolled. The main outcomes were the association between ICH detected solely by MRI and clinical outcomes at 90 days. Poor clinical outcomes were defined as mRS score > 2 at 90 days after onset. RESULTS: Of the 246 patients, 29 (12%) had ICH on MRI (MRI-ICH(+)), and 217 (88%) were MRI-ICH(-). There was no significant difference between number of patients with MRI-ICH(+) experiencing poor (10 [12%]) and favorable (19 [12%]) outcomes. The mRS score at 90 days between patients with MRI-ICH (+) and MRI-ICH(-) was not significantly different (2 [1-4] vs. 2 [1-4], respectively). Higher age and lower ASPECTS were independent risk factors for poor outcomes, as shown by multivariate regression analysis. MRI-ICH(+) status was not associated with poor outcomes. CONCLUSIONS: ICH detected by MRI alone did not influence clinical outcomes in patients with LVO treated with MT.

An efficient deep neural network for automatic classification of acute intracranial hemorrhages in brain CT scans.

BACKGROUND: Recent advancements in deep learning models have demonstrated their potential in the field of medical imaging, achieving remarkable performance surpassing human capabilities in tasks such as classification and segmentation. However, these modern state-of-the-art network architectures often demand substantial computational resources, which limits their practical application in resource-constrained settings. This study aims to propose an efficient diagnostic deep learning model specifically designed for the classification of intracranial hemorrhage in brain CT scans. METHOD: Our proposed model utilizes a combination of depthwise separable convolutions and a multi-receptive field mechanism to achieve a trade-off between performance and computational efficiency. The model was trained on RSNA datasets and validated on CQ500 dataset and PhysioNet dataset. RESULT: Through a comprehensive comparison with state-of-the-art models, our model achieves an average AUROC score of 0.952 on RSNA datasets and exhibits robust generalization capabilities, comparable to SE-ResNeXt, across other open datasets. Furthermore, the parameter count of our model is just 3 % of that of MobileNet V3. CONCLUSION: This study presents a diagnostic deep-learning model that is optimized for classifying intracranial hemorrhages in brain CT scans. The efficient characteristics make our proposed model highly promising for broader applications in medical settings.

Factors related to vessel displacement due to stent retriever retraction: An in vitro study.

BACKGROUND: Thrombectomy with a stent retriever (SR) may lead to intracranial hemorrhage due to vessel displacement. We aimed to explore factors related to vessel displacement using an in vitro vessel model. METHODS: A vessel model mimicking two-dimensional left internal carotid angiography findings was used in this study. Six SR types (Solitaire 3 × 40, 4 × 40, and 6 × 40; Embotrap 5 × 37; Trevo 4 × 41; and Tron 4 × 40) were fully deployed in the M2 ascending, M2 bend, or M1 horizontal portion. Subsequently, the SR was retracted, and the vessel displacement, maximum SR retraction force, and angle of the M2 bend portion were measured. A total of 180 SR retraction experiments were conducted using 6 SR types at 3 deployment positions with 10 repetitions each. RESULTS: The mean maximum distance of vessel displacement for Embotrap Ⅲ 5 × 37 (6.4 ± 3.5 mm, n = 30) was significantly longer than that for the other five SR types (p = 0.029 for Solitaire 6 × 40 and p < 0.001 for the others, respectively). Vessel displacement was significantly longer in the M2 ascending portion group (5.4 ± 3.0 mm, n = 60) than in the M2 bend portion group (3.3 ± 1.6 mm, n = 60) (p < 0.001) and it was significantly longer in the M2 bend portion group than in the M1 horizontal portion group (1.1 ± 0.7 mm, n = 60) (p < 0.001). A positive correlation existed between the mean maximum SR retraction force or mean angle of the M2 bend portion due to SR retraction (i.e., vessel straightening) and the mean maximum distance of vessel displacement (r = 0.90, p < 0.001; r = 0.90, p < 0.001, respectively). CONCLUSIONS: Vessel displacement varied with the SR type, size, and deployment position. Moreover, vessel displacement correlated with the SR retraction force or vessel straightening of the M2 bend portion.

Precision and Speed at Your Fingertips: An Automated Intracranial Hematoma Volume Calculation.

BACKGROUND: Intracranial hemorrhage (ICH) is a severe condition that requires rapid diagnosis and treatment. Automated methods for calculating ICH volumes can reduce human error and improve clinical decisioPlease provide professional degrees (e.g., PhD, MD) for the corresponding author.n-making. A novel automated method has been developed that is comparable to the ABC/2 method in terms of speed and accuracy while providing more accurate volumetric data. METHODS: We developed a novel automated algorithm for calculating intracranial blood volume from computed tomography (CT) scans. The algorithm consists of a Python script that processes Digital Imaging and Communications in Medicine images and determines the blood volume and ratio. The algorithm was validated against manual calculations performed by neurosurgeons. RESULTS: Our novel automated algorithm for calculating intracranial blood volume from CT scans demonstrated excellent agreement with the ABC/2 method, with a median overall difference of just 1.46 mL. The algorithm was also validated in patient groups with ICH, epidural hematoma (EDH), and SDH, with agreement coefficients of 0.992, 0.983, and 0.997, respectively. CONCLUSIONS: The study introduces a novel automated algorithm for calculating the volumes of various ICHs (EDH, and SDH) within CT scans. The algorithm showed excellent agreement with manual calculations and outperformed the commonly used ABC/2 method, which tends to overestimate ICH volume. The automated algorithm offers a more accurate, efficient, and time-saving approach to quantifying ICH, EDH, and SDH volumes, making it a valuable tool for clinical evaluation and decision-making.

The association between fetal intracranial hemorrhages detected on MRI and neurodevelopment.

PURPOSE: Fetal intracranial hemorrhage is rarely identified in prenatal imaging. When identified, sparse data regarding neurodevelopmental outcomes worsens prenatal dilemmas. This MRI-based study aimed to assess prenatal characteristics and neurodevelopmental outcomes of fetal intracranial hemorrhage. METHODS: A historical cohort study which identified fetal intracranial hemorrhage in 22 individual fetal MRI scans, as part of the assessment of abnormal prenatal sonographic findings. Severity was graded by the grading system commonly used in neonates, with modifications. Prenatal data was collected. Neurodevelopmental outcome was assessed clinically by Vineland-II Adaptive Behavior Scales. RESULTS: Eight fetuses had intraventricular hemorrhage grade I-II, twelve had intraventricular hemorrhage grade III-IV, and two had infratentorial hemorrhage. The most prevalent risk factors were maternal chronic diseases and chronic use of medications. There was male predominance. Pregnancy was terminated in eleven cases. No surviving child who participated in the Vineland assessment had a grade IV hemorrhage. Vineland scores were normal in 9/11 children and moderately low in 2/11. The mean composite score of the cohort was not different from the mean score expected for age. Clinically, one child had hypotonia. CONCLUSIONS: Prognosis for fetuses with ICH without parenchymal involvement is potentially more favorable than expected from the intraventricular hemorrhage grading-scale adopted from the preterm neonates. Parenchymal involvement may predict a worse outcome, but it is not the sole predicting feature. This information may be valuable during prenatal counseling.

Imaging of intracranial hemorrhage in photon counting computed tomography using virtual monoenergetic images.

PURPOSE: To determine the optimal virtual monoenergetic image (VMI) for detecting and assessing intracranial hemorrhage in unenhanced photon counting CT of the head based on the evaluation of quantitative and qualitative image quality parameters. METHODS: Sixty-three patients with acute intracranial hemorrhage and unenhanced CT of the head were retrospectively included. In these patients, 35 intraparenchymal, 39 intraventricular, 30 subarachnoidal, and 43 subdural hemorrhages were selected. VMIs were reconstructed using all available monoenergetic reconstruction levels (40-190 keV). Multiple regions of interest measurements were used for evaluation of the overall image quality, and signal, noise, signal-to-noise-ratio (SNR), and contrast-to-noise-ratio (CNR) of intracranial hemorrhage. Based on the results of the quantitative analysis, specific VMIs were rated by five radiologists on a 5-point Likert scale. RESULTS: Signal, noise, SNR, and CNR differed significantly between different VMIs (p < 0.001). Maximum CNR for intracranial hemorrhage was reached in VMI with keV levels > 120 keV (intraparenchymal 143 keV, intraventricular 164 keV, subarachnoidal 124 keV, and subdural hemorrhage 133 keV). In reading, no relevant superiority in the detection of hemorrhage could be demonstrated using VMIs above 66 keV. CONCLUSION: For the detection of hemorrhage in unenhanced CT of the head, the quantitative analysis of the present study on photon counting CT is generally consistent with the findings from dual-energy CT, suggesting keV levels just above 120 keV and higher depending on the location of the hemorrhage. However, on the basis of the qualitative analyses, no reliable statement can yet be made as to whether an additional VMI with higher keV is truly beneficial in everyday clinical practice.

Tenecteplase for Stroke at 4.5 to 24 Hours with Perfusion-Imaging Selection.

BACKGROUND: Thrombolytic agents, including tenecteplase, are generally used within 4.5 hours after the onset of stroke symptoms. Information on whether tenecteplase confers benefit beyond 4.5 hours is limited. METHODS: We conducted a multicenter, double-blind, randomized, placebo-controlled trial involving patients with ischemic stroke to compare tenecteplase (0.25 mg per kilogram of body weight, up to 25 mg) with placebo administered 4.5 to 24 hours after the time that the patient was last known to be well. Patients had to have evidence of occlusion of the middle cerebral artery or internal carotid artery and salvageable tissue as determined on perfusion imaging. The primary outcome was the ordinal score on the modified Rankin scale (range, 0 to 6, with higher scores indicating greater disability and a score of 6 indicating death) at day 90. Safety outcomes included death and symptomatic intracranial hemorrhage. RESULTS: The trial enrolled 458 patients, 77.3% of whom subsequently underwent thrombectomy; 228 patients were assigned to receive tenecteplase, and 230 to receive placebo. The median time between the time the patient was last known to be well and randomization was approximately 12 hours in the tenecteplase group and approximately 13 hours in the placebo group. The median score on the modified Rankin scale at 90 days was 3 in each group. The adjusted common odds ratio for the distribution of scores on the modified Rankin scale at 90 days for tenecteplase as compared with placebo was 1.13 (95% confidence interval, 0.82 to 1.57; P = 0.45). In the safety population, mortality at 90 days was 19.7% in the tenecteplase group and 18.2% in the placebo group, and the incidence of symptomatic intracranial hemorrhage was 3.2% and 2.3%, respectively. CONCLUSIONS: Tenecteplase therapy that was initiated 4.5 to 24 hours after stroke onset in patients with occlusions of the middle cerebral artery or internal carotid artery, most of whom had undergone endovascular thrombectomy, did not result in better clinical outcomes than those with placebo. The incidence of symptomatic intracerebral hemorrhage was similar in the two groups. (Funded by Genentech; TIMELESS ClinicalTrials.gov number, NCT03785678.).

AI assisted reader evaluation in acute CT head interpretation (AI-REACT): protocol for a multireader multicase study.

INTRODUCTION: A non-contrast CT head scan (NCCTH) is the most common cross-sectional imaging investigation requested in the emergency department. Advances in computer vision have led to development of several artificial intelligence (AI) tools to detect abnormalities on NCCTH. These tools are intended to provide clinical decision support for clinicians, rather than stand-alone diagnostic devices. However, validation studies mostly compare AI performance against radiologists, and there is relative paucity of evidence on the impact of AI assistance on other healthcare staff who review NCCTH in their daily clinical practice. METHODS AND ANALYSIS: A retrospective data set of 150 NCCTH will be compiled, to include 60 control cases and 90 cases with intracranial haemorrhage, hypodensities suggestive of infarct, midline shift, mass effect or skull fracture. The intracranial haemorrhage cases will be subclassified into extradural, subdural, subarachnoid, intraparenchymal and intraventricular. 30 readers will be recruited across four National Health Service (NHS) trusts including 10 general radiologists, 15 emergency medicine clinicians and 5 CT radiographers of varying experience. Readers will interpret each scan first without, then with, the assistance of the qER EU 2.0 AI tool, with an intervening 2-week washout period. Using a panel of neuroradiologists as ground truth, the stand-alone performance of qER will be assessed, and its impact on the readers' performance will be analysed as change in accuracy (area under the curve), median review time per scan and self-reported diagnostic confidence. Subgroup analyses will be performed by reader professional group, reader seniority, pathological finding, and neuroradiologist-rated difficulty. ETHICS AND DISSEMINATION: The study has been approved by the UK Healthcare Research Authority (IRAS 310995, approved 13 December 2022). The use of anonymised retrospective NCCTH has been authorised by Oxford University Hospitals. The results will be presented at relevant conferences and published in a peer-reviewed journal. TRIAL REGISTRATION NUMBER: NCT06018545.

Pathological changes in the lenticulostriate artery indicate the mechanisms leading to intracranial hemorrhage in Moyamoya disease: a case report.

This case report details the pathological findings of a vessel wall identified as the bleeding point for intracranial hemorrhage associated with Moyamoya disease. A 29-year-old woman experienced intracranial hemorrhage unrelated to hyperperfusion following superficial temporal artery-middle cerebral artery bypass surgery. A pseudoaneurysm on the lenticulostriate artery (LSA) was identified as the causative vessel and subsequently excised. Examination of the excised pseudoaneurysm revealed a fragment of the LSA, with a disrupted internal elastic lamina and media degeneration. These pathological findings in a perforating artery, akin to the circle of Willis, provide insights into the underlying mechanisms of hemorrhage in Moyamoya disease.

Cranial Point-of-Care Ultrasound for Neonatal Providers: A Feasibility Study.

OBJECTIVE: Despite strong evidence for its utility in clinical management and diagnosis of intracranial hemorrhage (ICH), the use of neonatal cranial point-of-care ultrasound (POCUS) has not been standardized in neonatal intensive care units (NICUs) in the United States. The primary aim of this study was to evaluate the feasibility of training NICU providers to perform cranial POCUS by tracking the quality of image acquisition following training. METHODS: Observational single-center cohort study of cranial POCUS images obtained by trained neonatal practitioners (attendings, fellows, and advanced practice providers) using a protocol developed by a radiologist and neonatologist. Exams were performed on infants born ≤1250 g and/or ≤30 weeks gestation within the first 3 days after birth. A survey to assess attitudes regarding cranial POCUS was given before each of three training sessions. Demographic and clinical data collection were portrayed with descriptive statistics. Metrics of image quality were assessed by a radiologist and sonographer independently. Analysis of trends in quality of POCUS images over time was performed using a multinomial Cochran-Armitage test. RESULTS: Eighty-two cranial POCUS scans were performed over a 2-year period. Infant median age at exam was 14 hours (IQR 7-22 hours). Metrics of image quality depicted quarterly demonstrated a significant improvement in depth (P = .01), gain (P = .048), and quality of anatomy images captured (P < .001) over time. Providers perceived increased utility and safety of cranial POCUS over time. CONCLUSION: Cranial POCUS image acquisition improved significantly following care team training, which may enable providers to diagnose ICH at the bedside.

Risk of intracranial hemorrhage in brain arteriovenous malformations: a systematic review and meta-analysis.

BACKGROUND AND OBJECTIVE: Brain arteriovenous malformations (bAVMs) carry a risk of hemorrhage. We aim to identify factors associated with subsequent hemorrhages. METHODS: Systematic searches were conducted across the ScienceDirect, Medline, and Cochrane databases. Assessed risk factors included bAVM size, bAVM volume, hemorrhage and seizure presentations, presence of deep venous drainage, deep-seated bAVMs, associated aneurysms, and Spetzler-Martin grade. Subgroup analyses were conducted on prior treatments, hemorrhage presentation, AVM size, and type of management. RESULTS: The meta-analysis included 8 cohort studies and 2 trials, with 4,240 participants. Initial hemorrhage presentation (HR 2.41; 95% CI 1.94-2.98; p < 0.001), any deep venous drainage (HR 1.52; 95% CI 1.09-2.13; p = 0.01), and associated aneurysms (HR 1.78; 95% CI 1.41-2.23; p < 0.001) increased secondary hemorrhage risk. Conversely, higher Spetzler-Martin grades (HR 0.77; 95% CI 0.68-0.87; p < 0.001) and larger malformation volumes (HR 0.87; 95% CI 0.76-0.99; p = 0.04) reduced risk. Subgroups showed any deep venous drainage in patients without prior treatment (HR 1.64; 95% CI 1.25-2.15; p < 0.001), bAVM > 3 cm (HR 1.79; 95% CI 1.15-2.78; p = 0.01), and multimodal interventions (HR 1.69; 95% CI 1.12-2.53; p = 0.01) increased risk. The reverse effect was found for patients initially presented without hemorrhage (HR 0.79; 95% CI 0.67-0.93; p = 0.01). Deep bAVM was a risk factor in > 3 cm cases (HR 2.72; 95% CI 1.61-4.59; p < 0.001) and multimodal management (HR 2.77; 95% CI 1.66-4.56; p < 0.001). Kaplan-Meier analysis revealed increased hemorrhage risk for initial hemorrhage presentation, while cumulative survival was higher in intervened patients over 72 months. CONCLUSION: Significant risk factors for bAVMs hemorrhage include initial hemorrhage, any deep venous drainage, and associated aneurysms. Deep venous drainage involvement is a risk factor in cases without prior treatment, those with bAVM > 3 cm, and cases managed with multimodal interventions. Deep bAVM involvement also emerges as a risk factor in cases > 3 cm and those managed with multimodal approaches.

Endovascular embolisation of posterior condylar canal dural arteriovenous fistula.

We describe a rare case of dural arteriovenous fistula (dAVF) of the posterior condylar canal in a man in his 30s who presented with recent onset headache and neck pain and subsequently acute intracranial haemorrhage. Radiological workup showed a medulla bridging vein draining dAVF of the right posterior condylar canal supplied by a meningeal branch of the right occipital artery. A dilated venous sac was seen compressing over cerebellar tonsil on the right side. There was acute haemorrhage in the posterior fossa and fourth ventricle. He was successfully managed with transarterial endovascular embolisation via a supercompliant balloon microcatheter without any complication. The balloon microcatheter effectively prevented reflux of the liquid embolic agent into the parent artery and vasa nervosa of lower cranial nerves.

Clinical and imaging predictors for hemorrhagic transformation of acute ischemic stroke after endovascular thrombectomy.

BACKGROUND AND PURPOSE: Hemorrhagic transformation (HT) is a common complication of endovascular thrombectomy (EVT) in patients with acute ischemic stroke (AIS). Our study aims to investigate the clinical and imaging predictors of HT and symptomatic intracranial hemorrhage (sICH) in patients who underwent EVT. METHODS: A retrospective analysis of 118 patients undergoing EVT for acute anterior circulation stroke was performed. Potential clinical and imaging predictors of all patients were collected and multivariate logistic regression was performed. The risk prediction system was constructed according to the multivariate logistic regression results. RESULTS: The incidence of HT and sICH after EVT were 46.6% and 15.3%, respectively. The multivariate logistic regression results showed that Alberta Stroke Program Early CT Score (ASPECTS) (p = .001, odds ratio [OR] = 0.367, 95% [confidence interval] CI, 0.201-0.670), collateral status (p<.001, OR = 0.117, 95% CI, 0.042-0.325), relative cerebral blood flow (CBF) ratio (p = .025, OR = 0.943, 95% CI, 0.895-0.993), and blood glucose on admission (p = .012, OR = 1.258, 95% CI, 1.053-1.504) were associated with HT. While for sICH, collateral circulation (p = .007, OR = 0.148, 95% CI, 0.037-0.589), ASPECTS (p = .033, OR = 0.510, 95% CI, 0.274-0.946), and blood glucose (p = .005, OR = 1.304, 95% CI, 1.082-1.573) were independent factors. The predictive model for HT after EVT was established, and the sensitivity and specificity of it were 90.9% and 79.4%, respectively, with the area under the curve of 90.0% (84.5%-95.4%). CONCLUSION: Collateral status, ASPECTS, relative CBF ratio, and blood glucose on admission were predictors for HT in AIS patients, while collateral status, ASPECTS, and blood glucose on admission were also predictors for sICH. In addition, the established predictive model showed good diagnostic value for prediction of HT after EVT.

mcTFI QSM MRI ABC/2 intracranial hemorrhage to noncontrast head CT volume measurement equivalence.

BACKGROUND/OBJECTIVES: Intracranial hemorrhage (ICH) volume assessment is an important part of patient management and is routinely obtained by non-contrast head CT (NCHCT) using the validated ABC/2 measurement method. Because conventional MRI imaging sequences demonstrate variability in ICH appearance, volumetric analyses for MRI bleed volume in a standardized manner using ABC/2 is not possible. The recently introduced multiecho-complex total field inversion quantitative susceptibility mapping (mcTFI QSM) MRI technique, which maps brain tissue susceptibility to both depict brain tissue structures and quantify tissue susceptibility, may provide a viable alternative. In this study we evaluated mcTFI QSM ABC/2 ICH volume assessment relative to NCHCT. METHODS: Patients with ICH who had undergone NCHCT and MRI brain scans within 48 h were recruited for this retrospective study. The ABC/2 method was applied to estimate the bleed volume for both NCHCT and MRI by a CAQ-certified neuroradiologist with 10 years of experience and a trained laboratory assistant. Results were analyzed via Bland-Altman (B-A) and linear regression. RESULTS: 54 patients (27 females) who had undergone NCHCT and MRI within 48 h (<24 h., n = 31, 24-48 h, n = 10) were enrolled. mcTFI QSM ICH volume measurement method showed a positive correlation (99.5%) compared to NCHCT. B-A plot comparing ABC/2 ICH volume on NCHCT and mcTFI MRI done for patients within 24 h demonstrates a bias of -0.09%. CONCLUSIONS: ICH volume calculation using ABC/2 on mcTFI QSM showed a high correlation with NCHCT measurement. These results suggest mcTFI QSM is a promising MRI method for ABC/2 for bleed volume measurement.

Examination-Level Supervision for Deep Learning-based Intracranial Hemorrhage Detection on Head CT Scans.

Purpose To compare the effectiveness of weak supervision (ie, with examination-level labels only) and strong supervision (ie, with image-level labels) in training deep learning models for detection of intracranial hemorrhage (ICH) on head CT scans. Materials and Methods In this retrospective study, an attention-based convolutional neural network was trained with either local (ie, image level) or global (ie, examination level) binary labels on the Radiological Society of North America (RSNA) 2019 Brain CT Hemorrhage Challenge dataset of 21 736 examinations (8876 [40.8%] ICH) and 752 422 images (107 784 [14.3%] ICH). The CQ500 (436 examinations; 212 [48.6%] ICH) and CT-ICH (75 examinations; 36 [48.0%] ICH) datasets were employed for external testing. Performance in detecting ICH was compared between weak (examination-level labels) and strong (image-level labels) learners as a function of the number of labels available during training. Results On examination-level binary classification, strong and weak learners did not have different area under the receiver operating characteristic curve values on the internal validation split (0.96 vs 0.96; P = .64) and the CQ500 dataset (0.90 vs 0.92; P = .15). Weak learners outperformed strong ones on the CT-ICH dataset (0.95 vs 0.92; P = .03). Weak learners had better section-level ICH detection performance when more than 10 000 labels were available for training (average f1 = 0.73 vs 0.65; P < .001). Weakly supervised models trained on the entire RSNA dataset required 35 times fewer labels than equivalent strong learners. Conclusion Strongly supervised models did not achieve better performance than weakly supervised ones, which could reduce radiologist labor requirements for prospective dataset curation. Keywords: CT, Head/Neck, Brain/Brain Stem, Hemorrhage Supplemental material is available for this article. © RSNA, 2023 See also commentary by Wahid and Fuentes in this issue.