Deep Learning-Based Automated Thrombolysis in Cerebral Infarction Scoring: A Timely Proof-of-Principle Study.

Background and Purpose: Mechanical thrombectomy is an established procedure for treatment of acute ischemic stroke. Mechanical thrombectomy success is commonly assessed by the Thrombolysis in Cerebral Infarction (TICI) score, assigned by visual inspection of X-ray digital subtraction angiography data. However, expert-based TICI scoring is highly observer-dependent. This represents a major obstacle for mechanical thrombectomy outcome comparison in, for instance, multicentric clinical studies. Focusing on occlusions of the M1 segment of the middle cerebral artery, the present study aimed to develop a deep learning (DL) solution to automated and, therefore, objective TICI scoring, to evaluate the agreement of DL- and expert-based scoring, and to compare corresponding numbers to published scoring variability of clinical experts. Methods: The study comprises 2 independent datasets. For DL system training and initial evaluation, an in-house dataset of 491 digital subtraction angiography series and modified TICI scores of 236 patients with M1 occlusions was collected. To test the model generalization capability, an independent external dataset with 95 digital subtraction angiography series was analyzed. Characteristics of the DL system were modeling TICI scoring as ordinal regression, explicit consideration of the temporal image information, integration of physiological knowledge, and modeling of inherent TICI scoring uncertainties. Results: For the in-house dataset, the DL system yields Cohen's kappa, overall accuracy, and specific agreement values of 0.61, 71%, and 63% to 84%, respectively, compared with the gold standard: the expert rating. Values slightly drop to 0.52/64%/43% to 87% when the model is, without changes, applied to the external dataset. After model updating, they increase to 0.65/74%/60% to 90%. Literature Cohen's kappa values for expert-based TICI scoring agreement are in the order of 0.6. Conclusions: The agreement of DL- and expert-based modified TICI scores in the range of published interobserver variability of clinical experts highlights the potential of the proposed DL solution to automated TICI scoring.

Comparison of acetylsalicylic acid and clopidogrel non-responsiveness assessed by light transmittance aggregometry and PFA-100® in patients undergoing neuroendovascular procedures.

Objectives: Dual platelet inhibition is commonly used for prevention of cardiovascular events in patients undergoing neuroendovascular procedures. Non-responsiveness to platelet inhibitors may be associated with adverse outcomes. The aim of this study was to evaluate the reliability of the platelet function analyzer PFA-100® in comparison to light transmittance aggregometry (LTA) for monitoring clopidogrel and acetylsalicylic acid (ASA) non-responsiveness in a cohort of patients treated for intracranial aneurysm or cranial artery stenosis. Methods: Non-responsiveness to clopidogrel and ASA was assessed by LTA using adenosine diphosphate (ADP) and arachidonic acid and by PFA-100® with the ADP/prostaglandin E1 (PGE1) and collagen/epinephrine cartridges, respectively. Results: A total of 203 patients (145 females; median age, 57 years) were analyzed. Agreement between the two tests was poor for clopidogrel non-responsiveness (ƙ=0.19) and not better than chance for ASA non-responsiveness (ƙ=0.01). Clopidogrel non-responsiveness by LTA and PFA-100® was associated with higher von Willebrand factor antigen and activity levels. ADP-induced platelet disaggregation was lower in patients with clopidogrel non-responsiveness as assessed by PFA-100®. Clopidogrel non-responsiveness by LTA was associated with a higher prevalence of diabetes and a higher body mass index (BMI). Adverse outcomes (death, thromboembolism, or in-stent thrombosis) occurred in 13% (n=26) of all patients independently of ASA and clopidogrel non-responsiveness as assessed by both devices. Conclusions: Our results show that LTA and PFA-100® are not interchangeable in the assessment of ASA and clopidogrel non-responsiveness in patients undergoing neuroendovascular interventions.

Brain magnetic resonance imaging in imported malaria.

BACKGROUND: Previous studies have documented a spectrum of brain magnetic resonance imaging (MRI) abnormalities in patients with cerebral malaria, but little is known about the prevalence of such abnormalities in patients with non-cerebral malaria. The aim of this study was to assess the frequency of brain MRI findings in returning travellers with non-cerebral malaria. METHODS: A total of 17 inpatients with microscopically confirmed Plasmodium falciparum non-cerebral malaria underwent structural brain MRI at 3.0 Tesla, including susceptibility-weighted imaging (SWI). Presence of imaging findings was recorded and correlated with clinical findings and parasitaemia. RESULTS: Structural brain abnormalities included a hyperintense lesion of the splenium on T2-weighted imaging (n = 3) accompanied by visible diffusion restriction (n = 2). Isolated brain microhaemorrhage was detected in 3 patients. T2-hyperintense signal abnormalities of the white matter ranged from absent to diffuse (n = 10 had 0-5 lesions, n = 5 had 5-20 lesions and 2 patients had more than 50 lesions). Imaging findings were not associated with parasitaemia or HRP2 levels. CONCLUSION: Brain MRI reveals a considerable frequency of T2-hyperintense splenial lesions in returning travellers with non-cerebral malaria, which appears to be independent of parasitaemia.

Time-resolved assessment of collateral flow using 4D CT angiography in large-vessel occlusion stroke.

OBJECTIVES: In acute stroke patients with large vessel occlusion, collateral blood flow affects tissue fate and patient outcome. The visibility of collaterals on computed tomography angiography (CTA) strongly depends on the acquisition phase, but the optimal time point for collateral imaging is unknown. METHODS: We analysed collaterals in a time-resolved fashion using four-dimensional (4D) CTA in 82 endovascularly treated stroke patients, aiming to determine which acquisition phase best depicts collaterals and predicts outcome. Early, peak and late phases as well as temporally fused maximum intensity projections (tMIP) were graded using a semiquantitative regional leptomeningeal collateral score, compared with conventional single-phase CTA and correlated with functional outcome. RESULTS: The total extent of collateral flow was best visualised on tMIP. Collateral scores were significantly lower on early and peak phase as well as on single-phase CTA. Collateral grade was associated with favourable functional outcome and the strength of this relationship increased from earlier to later phases, with collaterals on tMIP showing the strongest correlation with outcome. CONCLUSIONS: Temporally fused tMIP images provide the best depiction of collateral flow. Our findings suggest that the total extent of collateral flow, rather than the velocity of collateral filling, best predicts clinical outcome. KEY POINTS: • Collateral flow visibility on CTA strongly depends on the acquisition phase • tMIP offers the best visualisation of the extent of collaterals • Outcome prediction may be better with tMIP than with earlier phases.• Total extent of collaterals seems more important than their filling speed • If triggered too early, CTA may underestimate collateral flow.

Evaluation of intracranial stenting in a simulated training and assessment environment for neuroendovascular procedures.

Purpose: Given the inherent complexity of neurointerventional procedures and the associated risks of ionizing radiation exposure, it is crucial to prioritize ongoing training and improve safety protocols. The aim of this study is to assess a training and evaluation in-vitro environment using a vascular model of M1 stenosis, within a clinical angiography suite, without relying on animal models or X-ray radiation. Materials and methods: Using a transparent model replicating M1 stenosis, we conducted intracranial stenting procedures with four different setups (Gateway & Wingspan, Gateway & Enterprise, Neurospeed & Acclino, and Pharos Vitesse). A video camera was integrated with the angiography system's monitor for real-time visualization, while a foot switch was employed to simulate live fluoroscopy. Three neuroradiologists with varying levels of expertise performed each procedure for three times. The total duration of fluoroscopy as well as the time from passing the stenosis with the wire to completion of the procedure were recorded using a dedicated software designed for this experimental setup. Results: Compared to the Gateway & Wingspan procedure, the total fluoroscopy time reduced significantly with the Gateway & Enterprise, Neurospeed & Acclino, and Pharos Vitesse procedures by 51.56 s, 111.33 s, and 144.89 s, respectively (p < 0.001). Additionally, physicians with under 2 years and over 5 years of experience reduced FT by 62.83 s and 106.42 s, respectively, (p < 0.001), compared to a novice physician. Similar trends were noted for the time of wire distal to stenosis, with significant reductions for Neurospeed & Acclino and Pharos Vitesse compared to both Gateway & Wingspan as well as Gateway & Enterprise (all p < 0.001). Conclusion: Procedures requiring wire exchange maneuvers exhibited nearly twice the fluoroscopy time in comparison to balloon-mounted stenting or stent-placement via PTA balloon catheters. The more experienced neuroradiologist demonstrated significantly quicker performance in line with expectations in a real-life clinical setting, when compared to the less experienced interventionalist. This in-vitro setup allowed the evaluation of alternative technical approaches and differences in experience of operators without the use of animal models or X-ray. The setup combines advantages of simulators and silicone vessel models in a realistic working environment.

Angiographic reconstructions from whole-brain perfusion CT for the detection of large vessel occlusion in acute stroke.

BACKGROUND AND PURPOSE: Multimodal CT imaging consisting of nonenhanced CT, CT angiography (CTA), and whole-brain volume perfusion CT is increasingly used for acute stroke imaging. In these patients, presence of vessel occlusion is an important factor governing treatment decisions and possible endovascular therapy. The goal of this study was to assess the value and diagnostic accuracy of angiographic thin-slice volume perfusion CT reconstructions for the detection of intracranial large vessel occlusion in patients with stroke. METHODS: Fifty-eight patients with acute stroke received nonenhanced CT, CTA, and volume perfusion CT. All images were obtained on a 128-slice multidetector CT scanner. CT angiographic axial and coronal maximum-intensity projections of the head were reconstructed from conventional CTA and from the peak arterial scan of the volume perfusion CT data set (4-dimensional CTA). Images were assessed for the presence of intracranial vessel occlusion. The distribution of ischemic lesions was analyzed on perfusion parameter maps. RESULTS: On CTA, 30 patients (52%) had a total of 33 occluded intracranial artery segments. Twenty-eight occlusions were identified on 4-dimensional CTA, resulting in an 85% sensitivity with a positive predictive value of 97%. When combined with an analysis of the perfusion parameter maps, sensitivity of 4-dimensional CTA increased to 94% with a positive predictive value of 100%. CONCLUSIONS: In acute stroke, angiographic volume perfusion CT reconstructions may be a feasible option to detect intracranial arterial occlusion and evaluate patients for endovascular therapy. Sensitivity for detection of intracranial arterial occlusion can be increased by simultaneous assessment of perfusion parameter maps. Future studies should assess whether time-resolved 4-dimensional CTA may offer additional diagnostically relevant information compared with single-phase CTA.

Antegrade flow across incomplete vessel occlusions can be distinguished from retrograde collateral flow using 4-dimensional computed tomographic angiography.

BACKGROUND AND PURPOSE: In acute stroke patients with intracranial vessel occlusion, angiographic demonstration of antegrade contrast opacification distal to the occlusion site (termed the "clot outline sign") has been associated with higher rates of vessel recanalization. We sought to determine whether antegrade flow can be demonstrated on time-resolved 4-dimensional computed tomographic angiography (4-dimensional CTA), whether it can be distinguished from retrograde collateral flow, and if it can be used to predict early recanalization. METHODS: Fifty-seven acute stroke patients with intracranial anterior circulation vessel occlusion were retrospectively identified. All patients had received a multimodal computed tomography examination including thin-section 4-dimensional CTA and subsequent digital subtraction angiography as part of an endovascular procedure. Pretreatment 4-dimensional CTA and single-phase CTA were assessed for presence of antegrade contrast opacification distal to the occlusion site. Digital subtraction angiograms were reviewed for preintervention Thrombolysis in Cerebral Infarction grade, presence of the clot outline sign, as well as postintervention Thrombolysis in Cerebral Infarction grade. RESULTS: On 4-dimensional CTA, evidence of antegrade flow was present in 11 of 57 cases (19.3%). Compared with angiography, 4-dimensional CTA predicted antegrade flow with 100% sensitivity and 97.9% specificity. Single-phase CTA offered 40% sensitivity and 87.2% specificity. Early recanalization occurred in 3 patients (6.5%) after intravenous thrombolysis (n=46); all demonstrated antegrade flow on 4-dimensional CTA. CONCLUSIONS: Using 4-dimensional CTA, it is possible to noninvasively distinguish antegrade flow across a cerebral artery occlusion from retrograde collateral flow. Presence of antegrade flow on 4-dimensional CTA is associated with an increased chance of early vessel recanalization.

Activity-dependent site-specific changes of glutamate receptor composition in vivo.

The subunit composition of postsynaptic non-NMDA-type glutamate receptors (GluRs) determines the function and trafficking of the receptor. Changes in GluR composition have been implicated in the homeostasis of neuronal excitability and synaptic plasticity underlying learning. Here, we imaged GluRs in vivo during the formation of new postsynaptic densities (PSDs) at Drosophila neuromuscular junctions coexpressing GluRIIA and GluRIIB subunits. GluR composition was independently regulated at directly neighboring PSDs on a submicron scale. Immature PSDs typically had large amounts of GluRIIA and small amounts of GluRIIB. During subsequent PSD maturation, however, the GluRIIA/GluRIIB composition changed and became more balanced. Reducing presynaptic glutamate release increased GluRIIA, but decreased GluRIIB incorporation. Moreover, the maturation of GluR composition correlated in a site-specific manner with the level of Bruchpilot, an active zone protein that is essential for mature glutamate release. Thus, we show that an activity-dependent, site-specific control of GluR composition can contribute to match pre- and postsynaptic assembly.

Novel synthetic clot analogs for in-vitro stroke modelling.

PURPOSE: The increased demand for training of mechanical thrombectomy in ischemic stroke and development of new recanalization devices urges the creation of new simulation models both for training and device assessment. Clots properties have shown to play a role in procedural planning and thrombectomy device effectiveness. In this study, we analyzed the characteristics and applicability of completely synthetic, animal-free clots in the setting of an in-vitro model of mechanical thrombectomy for training and device assessment. METHODS: Synthetic clots based on agarose (n = 12) and silicone (n = 11) were evaluated in an in-vitro neurointervention simulation of mechanical thrombectomy with clot extraction devices. Calcified clots of mixed nature were simulated with addition of 3D printed structures. 9 clots were excluded due to insufficient vessel occlusion and failure to integrate with clot extraction device. Synthetic thrombi were characterized and compared using a categorical score-system on vessel occlusion, elasticity, fragmentation, adherence and device integration. RESULTS: Both agarose-based and silicone-based clots demonstrated relevant flow arrest and a good integration with the clot extraction device. Silicone-based clots scored higher on adherence to the vessel wall and elasticity. CONCLUSION: Selected synthetic clots can successfully be implemented in an in-vitro training environment of mechanical thrombectomy. The clots' different properties might serve to mimic fibrin-rich and red blood cell-rich human thrombi.

Comparing different MR angiography strategies of carotid stents in a vascular flow model: toward stent-specific recommendations in MR follow-up.

INTRODUCTION: Carotid artery stenting (CAS) requires adequate follow-up imaging to assess complications such as in-stent stenosis or occlusion. Options include digital subtraction angiography, CT angiography, ultrasound, and MR angiography (MRA), which may offer a non-invasive option for CAS follow-up imaging. The aim of this study was to assess contrast-enhanced MRA (CE-MRA) and three-dimensional time-of-flight MRA (3D-TOF) for visualization of the in-stent lumen in different carotid stents. METHODS: In this study, we compared CE-MRA and 3D-TOF of five different carotid stents (Guidant Acculink®, Cordis Precise®, Boston Wallstent®, Abbot Vascular Xact®, Cook Zilver®) in three diameters (4, 6, and 8 mm) using a vascular flow model at 3.0 T with the help of a recently developed carotid surface coil. Stent-related artifacts were objectively assessed by calculating artificial lumen narrowing (ALN) and relative in-stent signal (RIS). RESULTS: RIS and ALN depended heavily on stent type, stent diameter, and the employed MR sequence. ALN and RIS were relatively favorable for Acculink®, Precise®, and Zilver® stents with both CE-MRA and 3D-TOF. CE-MRA provided better results for the Wallstent, while the Xact stent was difficult to visualize with both MRA protocols. CONCLUSION: Both CE-MRA and 3D-TOF are viable options for depicting the in-stent lumen in carotid stents. For specific stents, 3D-TOF provided image quality comparable to CE-MRA and may thus be suitable for in vivo assessment. Development of stent-specific pathways for follow-up imaging seems advisable to address stent-related differences in image quality.

Unusual association of alveolar rhabdomyosarcoma with pancreatic metastasis: emerging role of PET-CT in tumor staging.

BACKGROUND: Pancreatic metastases in childhood cancer have been rarely reported in the radiology literature although ample evidence exists in pathology reports for its occurrence in patients with alveolar rhabdomyosarcomas (RMS). OBJECTIVE: Assess the occurrence of pancreatic metastases in alveolar rhabdomyosarcomas, increase awareness of this association and reassess current staging protocols. MATERIALS AND METHODS: Three major oncology centers reviewed their records and imaging examinations. Patients' history and demographics, primary tumor site and histology, presence of tumor recurrence, and presence and location of other metastases were reviewed. RESULTS: Pancreatic metastases occurred in eight patients with alveolar RMS. Four of these presented at diagnosis and four with disease recurrence. In recurrent disease, the duration between the diagnosis of the primary tumor and pancreatic metastases varied from 8 months to 6 years (mean +/- SD: 2.38 +/- 2.49 years). In all patients who received PET scans, pancreatic metastases showed a marked FDG-uptake, but had variable detectability with CT. Pancreatic metastases were not associated with certain primary tumor locations or presence of other metastases, mandating an evaluation of the pancreas in all cases of alveolar rhabdomyosarcomas. CONCLUSION: Radiologists should be sensitized and actively evaluate the pancreas in patients with alveolar RMS. Optimizing CT and PET-CT protocols may increase the diagnostic yield.

Confidence of treatment decision and perceived risk of procedure-related neurological complications in the management of unruptured intracranial aneurysms.

BACKGROUND AND PURPOSE: To evaluate factors influencing the confidence of management recommendation for unruptured intracranial aneurysms (UIAs) and to assess the ability of neurointerventionalists to predict procedure-related neurological complications compared with a 3-point risk score. MATERIALS AND METHODS: Twenty-eight neurointerventionalists were asked to evaluate digital subtraction angiographies examinations of patients with UIAs by determining the best management approach, their level of confidence in their management recommendation, and estimating the risk of procedure-related neurological complications. Knowledge and experience in interventional neuroradiology (INR) of each participant were assessed. RESULTS: Reliability was moderate regarding any treatment recommendation (ICC=0.49) and low regarding the estimation of risk of complications (ICC=0.38). The recommendation of clipping was less likely with more experience in INR (OR=0.6) and more likely with increasing knowledge (OR=1.7). Odds of recommending WEB device were lower with more experience in INR (OR=0.6), higher in patients with multiple aneurysms (OR=3.6) and increasing neck width (OR=2.7). The recommendation of stent-assisted coiling was more likely with increasing neck width (OR=2.4) and when cerebral ischemic comorbidities were present (OR=2.9). The participants were significantly worse than the risk score (mean area under the curve of 0.53) and not better than random guess in predicting complications. Neither knowledge nor experience in INR was significantly associated with the participants' ability to predict neurological complications. CONCLUSIONS: Our study shows a moderate interrater reliability of treatment recommendations of UIAs. Confidence in treatment recommendation varied significantly according to recommended treatments. Overall performance in predicting neurological complications was worse than the risk score and not better than random guess.

Voxel-Based Sensitivity of Flat-Panel CT for the Detection of Intracranial Hemorrhage: Comparison to Multi-Detector CT.

OBJECTIVES: Flat-panel CT (FPCT) allows cross-sectional parenchymal, vascular and perfusion imaging within the angiography suite, which could greatly facilitate acute stroke management. We hypothesized that FPCT offers equal diagnostic accuracy compared to multi-detector CT (MDCT) as a primary tool to exclude intracranial hemorrhage. METHODS: 22 patients with intracranial hematomas who had both MDCT and FPCT performed within 24 hours were retrospectively identified. Patients with visible change in hematoma size or configuration were excluded. Two raters independently segmented hemorrhagic lesions. Data sets and corresponding binary lesion maps were co-registered to compare hematoma volume. Diagnostic accuracy of FPCT to detect hemorrhage was calculated from voxel-wise analysis of lesion overlap compared to reference MDCT. RESULTS: Mean hematoma size was similar between MDCT (16.2±8.9 ml) and FPCT (16.1±8.6 ml), with near perfect correlation of hematoma sizes between modalities (ρ = 0.95, p<0.001). Sensitivity and specificity of FPCT to detect hemorrhagic voxels was 61.6% and 99.8% for intraventricular hematomas and 67.7% and 99.5% for all other intracranial hematomas. CONCLUSIONS: In this small sample containing predominantly cases with subarachnoid hemorrhage, FPCT based assessment of hemorrhagic volume in brain yields acceptable accuracy compared to reference MDCT, albeit with a limited sensitivity on a voxel level. Further assessment and improvement of FPCT is necessary before it can be applied as a primary imaging modality to exclude intracranial hemorrhage in acute stroke patients.

Quantitative Evaluation of Performance in Interventional Neuroradiology: An Integrated Curriculum Featuring Theoretical and Practical Challenges.

PURPOSE: We sought to develop a standardized curriculum capable of assessing key competencies in Interventional Neuroradiology by the use of models and simulators in an objective, quantitative, and efficient way. In this evaluation we analyzed the associations between the practical experience, theoretical knowledge, and the skills lab performance of interventionalists. MATERIALS AND METHODS: We evaluated the endovascular skills of 26 participants of the Advanced Course in Endovascular Interventional Neuroradiology of the European Society of Neuroradiology with a set of three tasks (aneurysm coiling and thrombectomy in a virtual simulator and placement of an intra-aneurysmal flow disruptor in a flow model). Practical experience was assessed by a survey. Participants completed a written and oral examination to evaluate theoretical knowledge. Bivariate and multivariate analyses were performed. RESULTS: In multivariate analysis knowledge of materials and techniques in Interventional Neuroradiology was moderately associated with skills in aneurysm coiling and thrombectomy. Experience in mechanical thrombectomy was moderately associated with thrombectomy skills, while age was negatively associated with thrombectomy skills. We found no significant association between age, sex, or work experience and skills in aneurysm coiling. CONCLUSION: Our study gives an example of how an integrated curriculum for reasonable and cost-effective assessment of key competences of an interventional neuroradiologist could look. In addition to traditional assessment of theoretical knowledge practical skills are measured by the use of endovascular simulators yielding objective, quantitative, and constructive data for the evaluation of the current performance status of participants as well as the evolution of their technical competency over time.

Common genomic aberrations in basaloid squamous cell carcinoma and carcinosarcoma of the esophagus detected by CGH and array CGH.

Basaloid squamous cell carcinoma (BSCC) and carcinosarcoma of the esophagus are rare entities, making up fewer than 2% of esophageal malignancies. Comparative genomic hybridization (CGH) in 1 case of BSCC and 2 cases of carcinosarcoma and subsequent array CGH in 1 case each of BSCC and carcinosarcoma revealed common chromosomal gains at 2p25.3-2p12, 7q21.3-7q22.3, and 11q13.2-11q13.4. Chromosomal losses at 13q31qter were observed in both carcinosarcomas. In addition, progression of genomic instability from in situ to invasive carcinosarcoma could be demonstrated by using array CGH. Our observations suggest a common genetic origin of BSCC and carcinosarcoma.