Search

Cross-modal retrieval(CMR) refers to the task of retrieving semantically related items across different modalities. For example, given an image query, the task is to retrieve relevant text descriptions or audio clips. One of the major challenges in CMR is the modality gap, which refers to the differences between the features and representations used to encode information in different modalities. To address the modality gap, researchers have developed various techniques such as joint embedding, where the features from different modalities are mapped to a common embedding space where they can be compared directly. Binary-valued and real-valued representations are two different ways to represent data. A binary-valued representation is a type of discrete representation where data is represented using either 0 or 1. Real-valued representation, on the other hand, represents each item as a vector of real numbers. Both types of representations have their advantages and disadvantages, and researchers continue to explore new techniques for generating representations that can improve the performance of CMR systems. First time, the work presented here generates both the representations and comparison is made by performing experiments on standard benchmark datasets using mean average precision (MAP). The result suggest that real-valued representation outperforms binary-valued representation in terms of MAP, especially when the data is complex and high-dimensional. On the other hand, binary codes are more memory-efficient than real-valued embedding, and they can be computed much faster. Moreover, binary codes can be easily stored and transmitted, making them more suitable for large-scale retrieval tasks.

Background: Understanding clot composition may allow for better technical planning and improved outcomes in mechanical thrombectomy (MT). We sought to correlate clot composition with reperfusion outcomes in MT. Methods: Clinical, imaging, and clot data from patients enrolled in the STRIP Registry from September 2016 to September 2020 were analyzed. Samples were fixed in 10% phosphate-buffered formalin and stained with hematoxylin and eosin and Martius Scarlett Blue. Percent composition, richness, and gross appearance were evaluated. Outcome measures included the rate of First-Pass Effect (mTICI2c/3, FPE) and the number of device passes. Results: A total of 1430 patients (mean age, 68.4±13.5years; median [IQR] baseline NIHSS,17.2 [10.5-23]; IV-tPA use, 36%; Stent-Retrievers [SR], 27%; Contact Aspiration [CA], 27%; Combined SR + CA, 43%) were included. The median [IQR] number of passes was 1 [1-2]. FPE was achieved in 39.3% of the cases. There was no association between percent histological composition or clot richness and FPE in the overall population. However, the combined technique resulted in lower FPE rates for RBC-rich (P Conclusion: Despite the lack of correlation between clot histology and FPE in the overall population, our study adds to the growing body of evidence supporting the notion that clot composition influences reperfusion treatment strategy outcomes. Additional studies are needed.

Background and Purpose: Despite the lower rates of good outcomes and higher mortality in elderly patients, age does not modify the treatment effect of mechanical thrombectomy for large vessel occlusion strokes. We aimed to study whether racial background influences the outcome after mechanical thrombectomy in the elderly population. Methods: We reviewed a prospectively maintained database of patients with acute ischemic stroke treated with mechanical thrombectomy from October 2010 through June 2020 to identify all consecutive patients with age ≥80 years and anterior circulation large vessel occlusion strokes. The patients were categorized according to their race as Black and White. Univariable and multivariable analyses were performed to define the predictors of 90-day modified Rankin Scale and mortality in the overall population and in each race separately. Results: Among 2241 mechanical thrombectomy, a total of 344 patients (median [interquartile range]; age 85 [82–88] years, baseline National Institutes of Health Stroke Scale score of 19 [15–23], Alberta Stroke Program Early CT Score 9 [7–9], 69.5% females) were eligible for the analysis. White patients (n=251; 73%) had significantly lower median body mass index (25.37 versus 26.89, P =0.04) and less frequent hypertension (78.9% versus 90.3%, P =0.01) but more atrial fibrillation (64.5% versus 44.1%, P =0.001) compared with African Americans (n=93; 27%). Other clinical, imaging, and procedural characteristics were comparable between groups. The rates of symptomatic intracerebral hemorrhage, 90-day modified Rankin Scale score of 0 to 2, and mortality were comparable among both groups. On multivariable analysis, race was neither a predictor of 90-day modified Rankin Scale score of 0 to 2 (White race: odds ratio, 0.899 [95% CI, 0.409–1.974], P =0.79) nor 90-day mortality (White race: odds ratio, 1.368; [95% CI, 0.715–2.618], P =0.34). Conclusions: In elderly patients undergoing mechanical thrombectomy for acute ischemic stroke, there was no racial difference in terms of outcome.

Introduction: Expediting notification of lesions in acute ischemic stroke (AIS) is critical. Limited availability of experts to assess such lesions and delays in large vessel occlusion (LVO) recognition can negatively affect outcomes. Artificial intelligence (AI) may aid LVO recognition and treatment. This study aims to evaluate the performance of an AI-based algorithm for LVO detection in AIS. Methods: Retrospective analysis of a database of AIS patients admitted in a single center between 2014 and 2019. Vascular neurologists graded computed tomography angiographies (CTAs) for presence and site of LVO. Studies were analyzed by the Viz-LVO Algorithm® version 1.4 – neural network programmed to detect occlusions from the internal carotid artery terminus (ICA-T) to the Sylvian fissure. Comparisons between human versus AI-based readings were done by test characteristic analysis and Cohen’s kappa. Primary analysis included ICA-T and/or middle cerebral artery (MCA)-M1 LVOs versus non-LVOs/more distal occlusions. Secondary analysis included MCA-M2 occlusions. Results: 610 CTAs were analyzed. The AI algorithm rejected 2.5% of the CTAs due to poor quality, which were excluded from the analysis. Viz-LVO identified ICA-T and MCA-M1 LVOs with a sensitivity of 87.6%, specificity of 88.5%, and accuracy of 87.9% (AUC 0.88, 95% CI: 0.85–0.92, p < 0.001). Cohen’s kappa was 0.74. In the secondary analysis, the algorithm yielded a sensitivity of 80.3%, specificity of 88.5%, and accuracy of 82.7%. The mean run time of the algorithm was 2.78 ± 0.5 min. Conclusion: Automated AI reading allows for fast and accurate identification of LVO strokes with timely notification to emergency teams, enabling quick decision-making for reperfusion therapies or transfer to specialized centers if needed.

Background and Purpose: Field Assessment Stroke Triage for Emergency Destination (FAST-ED) scale is a helpful tool to triage patients with stroke in the field. However, data on its reliability in the prehospital setting are lacking. We aim to test the reliability of FAST-ED scale when used by paramedics in a mobile stroke unit covering a metropolitan area. Methods: As part of standard operating mobile stroke unit procedures, paramedics initially evaluated patients. If the event characterized a stroke alert, the FAST-ED score was determined by the paramedic upon patient contact (in-person) and then independently by a vascular neurologist (VN) immediately after paramedic evaluation (remotely/telemedicine). This allowed testing of the interrater agreement of the FAST-ED scoring performance between on-site prehospital providers and remotely located VN. Results: Of a total of 238 patients transported in the first 15 months of the mobile stroke unit’s activity, 173 were included in this study. Median age was 63 (interquartile range, 55.5–75) years and 52.6% were females. A final diagnosis of ischemic stroke was made in 71 (41%), transient ischemic attack in 26 (15%), intracranial hemorrhage in 15 (9%), whereas 61 (35%) patients were stroke mimics. The FAST-ED scores matched perfectly among paramedics and VN in 97 (56%) instances, while there was 0 to 1-point difference in 158 (91.3%), 0 to 2-point difference in 171 (98.8%), and 3 or more point difference in 2 (1.1%) patients. The intraclass correlation between VN and paramedic FAST-ED scores showed excellent reliability, intraclass correlation coefficient 0.94 (95% CI, 0.92–0.96; P Conclusions: We demonstrate excellent reliability of FAST-ED scale performed by paramedics when compared with VN, indicating that it can be accurately performed by paramedics in the prehospital setting.

Background: WSN is a network of smart tiny electromechanical devices named sensors. Sensors perform various tasks like sensing the environment as per its range, transmitting the data using transmission units, storing the data in the storage unit and performing an action based on the captured data. As they are installed in an unfriendly environment, recharging the sensors is not possible every time which leads to a limited lifetime of the network. To enhance the life of a sensor network, the network requires energy-efficient protocols. Various energy-efficient MAC protocols have been developed by the research community, but very few of them are integrated with the priority-based environment which performs priority-based data transmission. Another challenge of WSN is that most of the WSN areas are delay-sensitive because it is implemented in critical fields like military, disaster management, and health monitoring. Energy, delay, and throughput are major quality factors that affect the sensor network. Objective: In this paper, the aim is to design and develop a MAC Protocol for a field like the military where the system requires energy efficiency and priority-based data transmission. Method: In the proposed model, a cluster-based network with priority queues is formed that can achieve higher power efficiency and less delay for sensitive data. Results: In this research, the simulation of proposed MAC, TMAC and SMAC is done with different numbers of nodes, same inter-packet intervals, and variant inter-packet intervals. Based on the script simulation, result graphs are generated. Conclusion: The proposed work achieves greater lifetime compared to TMAC and SMAC using priority-based data transmission.

COVID-19 (Coronavirus disease 2019) caused by SARS-CoV-2 has a diverse constellation of neurological manifestations that include encephalopathy, stroke, Guillain–Barré syndrome, myelitis, and encephalitis. Intraluminal carotid thrombi (ILT) are infrequent lesions seen in only 1.6% of patients with acute ischemic stroke. Underlying atherosclerosis is the most common lesion associated with ILT formation. However, with COVID-19, we have encountered ILT in patients without significant atherosclerotic disease. The endothelial inflammation and hypercoagulable state associated with COVID-19 pose a risk of arterial and venous thromboembolism and could have contributed to this presentation although the exact pathophysiology and optimal treatment of ILT in COVID-19 remain elusive. Herein, we present a series of ischemic stroke patients with carotid ILT in the setting of a recent SARS-CoV-2 infection.

Reactive species (RS) are known to play significant roles in cancer development as well as in treating or managing cancer. On the other hand, genome scale metabolic models are being used to understand cell metabolism in disease contexts including cancer, and also in planning strategies to handle diseases. Despite their crucial roles in cancers, the reactive species have not been adequately modeled in the genome scale metabolic models (GSMMs) when probing disease models for their metabolism or detection of drug targets. In this work, we have developed a module of reactive species reactions, which is scalable - it can be integrated with any human metabolic model as it is, or with any metabolic model with fine-tuning. When integrated with a cancer (colorectal cancer in this case) metabolic model, the RS module highlighted the deregulation occurring in important CRC pathways such as fatty acid metabolism, cholesterol metabolism, arachidonic acid and eicosanoid metabolism. We show that the RS module helps in better deciphering crucial metabolic targets for devising better therapeutics such as FDFT1, FADS2 and GUK1 by taking into account the effects mediated by reactive species during colorectal cancer progression. The results from this reactive species integrated CRC metabolic model reinforces ferroptosis as a potential target for colorectal cancer therapy.

BackgroundThe optimal anesthesia management for patients with stroke undergoing mechanical thrombectomy (MT) during the COVID-19 pandemic has become a matter of controversy. Some recent guidelines have favored general anesthesia (GA) in patients perceived as high risk for intraprocedural conversion from sedation to GA, including those with dominant hemispheric occlusions/aphasia or baseline National Institutes of Health Stroke Scale (NIHSS) score >15. We aim to identify the rate and predictors of conversion to GA during MT in a high-volume center where monitored anesthesia care (MAC) is the default modality.MethodsA retrospective review of a prospectively maintained MT database from January 2013 to July 2020 was undertaken. Analyses were conducted to identify the predictors of intraprocedural conversion to GA. In addition, we analyzed the GA conversion rates in subgroups of interest.ResultsAmong 1919 MT patients, 1681 (87.6%) started treatment under MAC (median age 65 years (IQR 55–76); baseline NIHSS 16 (IQR 11–21); 48.4% women). Of the 1677 eligible patients, 26 (1.6%) converted to GA including 1.4% (22/1615) with anterior and 6.5% (4/62) with posterior circulation strokes. The only predictor of GA conversion was posterior circulation stroke (OR 4.99, 95% CI 1.67 to 14.96, P=0.004). The conversion rates were numerically higher in right than in left hemispheric occlusions (1.6% vs 1.2%; OR 1.37, 95% CI 0.59 to 3.19, P=0.47) and in milder than in more severe strokes (NIHSS ≤15 vs >15: 2% vs 1.2%; OR 0.62, 95% CI 0.28 to 1.36, P=0.23).ConclusionsOur study showed that the overall rate of conversion from MAC to GA during MT was low (1.6%) and, while higher in posterior circulation strokes, it was not predicted by either hemispheric dominance or stroke severity. Caution should be given before changing clinical practice during moments of crisis.

BackgroundFirst pass reperfusion (FPR) has been established as a key performance metric in mechanical thrombectomy (MT). The impact of FPR may be more relevant in fast progressors. We aim to study the impact of baseline Alberta Stroke Program Early CT Score (ASPECTS) on non-contrast CT and hypoperfusion intensity ratio (HIR) on CT perfusion on clinical outcomes after FPR.MethodsA prospective MT database was reviewed for patients with isolated occlusion of the intracranial internal carotid artery and/or middle cerebral artery M1 segment who underwent MT with complete reperfusion (modified Thrombolyis in Cerebral Infarction score 2c–3) from January 2012 to May 2019. The overall population was divided into ASPECTS >7 versus ≤7 and the subgroup of patients with baseline CT perfusion was divided into HIR ResultsA total of 436 patients were included in the analyses. FPR was achieved in 254 (58.3%) patients. ASPECTS modified the effect of FPR on clinical outcomes, with FPR predicting good outcomes in patients with ASPECTS ≤7 (46% vs 29%, adjusted OR 3.748; 95% CI 1.590 to 8.838, p=0.003) while no significant effect was detected in those with ASPECTS >7 (62.3% vs 53.1%, adjusted OR 1.372; 95% CI 0.798 to 2.358, p=0.25). Similarly, FPR predicted good outcomes in patients with HIR ≥0.3 (54.8% vs 41.9%, adjusted OR 2.204; 95% CI 1.148 to 4.233, p=0.01) but not in those with HIR ConclusionsThe impact of FPR on functional outcomes is highly dependent on baseline imaging characteristics, with a more prominent influence in patients presenting with lower ASPECTS and/or higher HIR.

Large scale synthesis of nanoparticles requires their simultaneous synthesis and separation. In this work, we establish a framework for simultaneous synthesis and separation of nanoparticles based ...

Introduction: During the initial peak of the COVID-19 pandemic, many centers globally reported a significant decrease in volumes of emergencies including acute stroke (AS) and acute myocardial infarction (AMI). While the reason for this remains unknown, pandemic-driven anxiety among patients may have resulted in unwarranted refusals to transport when deemed necessary by EMS (Emergency Medical Services) providers. We sought to study the impact of COVID-19 pandemic on the patterns of Emergency Medical transport (EMTr) and patient refusals to transport when serious medical conditions were suspected by EMS personnel. Methods: In this retrospective, observational study of Grady Health System’s EMS, we compared the rates of EMTr and refusals for adult patients with suspected diagnoses of AS, AMI, and other medical conditions in the first year of the pandemic (Y1, Mar 2020-Feb 2021) with the corresponding period in the year prior (Y0). We also compared the temporal trends for these variables across the different pandemic waves (1 st , Mar-May 2020; 2 nd , Jun-Aug 2020; 3 rd , Sep 2020-Feb 2021) with the corresponding periods in the year before. Results: Grady EMS responded to 207,888 calls in Y1 compared to 201,968 in Y0. The overall rate of refusals for all diagnoses was 15.5% in Y1 vs 14.1% in Y0, that for AS was 2.25% in Y1 vs 1.77% in Y0 and 7.5% in Y1 vs 5.67% in Y0 for AMI (Figure). Conclusion: There were more refusals in the first two waves of the pandemic. While refusals were higher for AS and AMI, this was not statistically significant. Our study provides valuable insight into the behavioral patterns of patients seeking emergency care during the pandemic and emphasizes a need for public education and more research.

Link prediction between nodes is an important problem in the study of complex networks. In this work, we investigate determining directed links in conserved flow networks from data. A novel approach to predict the unknown links of an underlying conserved network, and the directions of flows associated with them are proposed. The directed link prediction is achieved by finding the missing values in the incidence matrix of the network by combining singular value decomposition, principal component analysis and tools from graph theory using steady-state flow data. In contrast to traditional integer optimization-based approaches, the proposed approach accomplishes the task of directed link prediction in cubic time. The methodology is corroborated via synthetic studies for flow data generated from networks based on Erdős–Rényi, Watts–Strogatz and Barabási–Albert network models at different signal-to-noise ratios. It is empirically shown that all unknown directed links can be correctly predicted with large enough noisy data, in an asymptotic sense, to correctly estimate the missing values in the incidence matrix.

Background and Purpose— We aimed to evaluate the current practice patterns, safety and outcomes of patients who receive endovascular therapy (EVT) having mild neurological symptoms. Methods— From Jan 2010 to Jan 2018, 127,794 ischemic stroke patients were enrolled in the Florida-Puerto Rico Stroke Registry. Patients presenting within 24 hours of symptoms who received EVT were classified into mild (National Institutes of Health Stroke Scale [NIHSS] ≤5) or moderate/severe (NIHSS>5) categories. Differences in clinical characteristics and outcomes were evaluated using multivariable logistic regression. Results— Among 4110 EVT patients (median age, 73 [interquartile range=20] years; 50% women), 446 (11%) had NIHSS ≤5. Compared with NIHSS >5, those with NIHSS ≤5 arrived later to the hospital (median, 138 versus 101 minutes), were less likely to receive intravenous alteplase (30% versus 43%), had a longer door-to-puncture time (median, 167 versus 115 minutes) and more likely treated in South Florida (64% versus 53%). In multivariable analysis younger age, private insurance (versus Medicare), history of hypertension, prior independent ambulation and hospital size were independent characteristics associated with NIHSS ≤5. Among EVT patients with NIHSS ≤5, 76% were discharged home/rehabilitation and 64% were able to ambulate independently at discharge as compared with 53% and 32% of patients with NIHSS >5. Symptomatic intracerebral hemorrhage occurred in 4% of mild stroke EVT patients and 6.4% in those with NIHSS >5. Conclusions— Despite lack of evidence-based recommendations, 11% of patients receiving EVT in clinical practice have mild neurological presentations. Individual, hospital and geographic disparities are observed among endovascularly treated patients based on the severity of clinical symptoms. Our data suggest safety and overall favorable outcomes for EVT patients with mild stroke.

Processing of unordered and unbounded data is the prime requirement of the current businesses. Large amount of rapidly generated data demands the processing of the same without the storage and as per the timestamp associated with it. It is difficult to process these unbounded data with batch engine as the existing batch systems suffer from the delay intrinsic by accumulating entire incoming records in a group prior to process it. However windowing can be useful when dealing with unbounded data which pieces up a dataset into fixed chunks for processing with repeated runs of batch engine. Contrast to batch processing, stream handling system aims to process information that is gathered in a little timeframe. In this way, stream data processing ought to be coordinated with the flow of data. In the real world the event time is always skewed with the processing time which introduce issues of delay and completeness in incoming stream of data. In this paper, we presented the analysis on the watermark and trigger approach which can be used to manage these unconventional desires in the processing of unbounded data.

The quantitative understanding of the dissolution of nuclear fuel materials is essential for the process design and development of an industrial-scale nuclear fuel reprocessing plant. The main objective of this review article is to analyze the published data related to the dissolution of important nuclear materials, namely, urania, plutonia, thoria, and their oxides in the existing literature. The published results on rate-controlling step and reaction mechanism of dissolution processes are reconciled and reviewed in this work. Clear suggestions are made for future research work for the identification of rate-controlling step. Suggestions are also provided to overcome the shortfalls in the published data for the identification of intrinsic kinetics and mass-transfer rates.

Extraction of pure species spectra from mixture spectra is important in characterizing unknown mixtures as well as in the monitoring of chemical reactions. In many designed experimental studies, the mixture concentrations are completely known and partial knowledge of the spectra of some of species in a mixture may also be available. In this study, we extend the methods of ordinary least squares (OLS), principal component regression (PCR), and non-negative matrix factorization (NMF) to incorporate such additional information. The performances of the three proposed methods are evaluated using simulated and experimental data. Among these, the proposed constrained NMF (cNMF) method is shown to be best-suited for obtaining feasible and accurate estimates of pure species spectra from mixture spectra.

Continuous flow reactors such as plug flow reactors (PFRs) have been used in the identification of reaction systems (Moore, J. S.; Jensen, K. F. Angew. Chem. 2014, 126, 480−483). It is important to understand the impact of operating conditions, measurement strategies, and kinetic model identification methods on the identification of reaction systems in a PFR. Typically, for kinetic model identification using the simultaneous identification approach, concentrations are measured as a function of residence time under isothermal operating conditions in a PFR. On the other hand, concentrations and temperature are required to be measured as a function of residence time under non-isothermal conditions for kinetic model identification of multiple reaction systems in a PFR. In this work, we extend the incremental method for identification of kinetic and heat transport models using measurements from non-isothermal steady-state PFRs. This work investigates the role of experimental conditions, measurement strategies ...

In this work, we experimentally demonstrate the synthesis of different bimetallic Ag–Cu nanostructures like core–shell, Janus particles. It is shown that different nanoparticle structures are associated with different reaction pathways. The pathway is manipulated by modifying the operating conditions (reactant sequence in this case) of the reactions. Ascorbic acid and cetyltrimethylammonium bromide (CTAB) is used as the reducing agent and the capping agent, respectively. Silver nitrate and copper nitrate are used as precursors. The reaction is carried out under microwave-assisted heating, which intensifies metal ion reduction. Two protocols are studied: (I) sequential and (II) simultaneous addition of precursors. By changing the sequence of the precursor addition, the morphology of the bimetallic nanoparticles was altered. Janus particles of size 25–30 nm were formed under the simultaneous addition and the sequential addition when the silver precursor is added first, followed by the copper precursor. When the sequence of precursor addition was reversed, i.e., copper precursor followed by silver, Cu core-Ag shell particles of size 65 nm were formed. We propose mechanisms that help understand the formation of these different structures. Resistance to oxidation studies showed that the particles with Cu core-Ag shell morphology were more resistant to oxidation than Janus particles.

Background and Purpose: There is a robust relationship between the duration of ischemia and functional outcomes after mechanical thrombectomy. Higher number of mechanical thrombectomy passes strongly correlate with lower chances of favorable outcomes. Indeed, previous studies have suggested that after multiple passes the procedure may be futile. However, using uncontrolled thresholds to define thrombectomy futility might be misleading. We aim to compare the outcome of successful reperfusion after 4 to 5 passes and ≥6 passes with those of failed reperfusion. Methods: A prospectively acquired mechanical thrombectomy database from January 2012 to October 2019 was reviewed. Patients were included if they had intracranial internal carotid artery or middle cerebral artery-M1/M2 occlusions and either achieved successful reperfusion after ≥4 passes or failed reperfusion. Reperfused patients (mTICI2b-3) were divided into 2 subgroups; (1) 4 to 5 passes and (2) ≥6 passes. Each subgroup was compared with a matched group of mechanical thrombectomy failure (mTICI0-2a). The primary outcome was the shift in the degree of disability at 90-day as measured by the modified Rankin Scale. Results: A total of 273 patients were included. As compared with matched failed reperfusion patients (n=62), those reperfused after 4 to 5 passes (n=62) had a favorable shift in the overall modified Rankin Scale score distribution (adjusted odds ratio, 3.992 [95% CI, 1.807–8.512], P =0.001] and higher rates of functional independence (31% versus 8.9%, P =0.004, adjusted odds ratio; 9.860 [95% CI, 2.323–41.845], P =0.002) at 90 days. Similarly, when compared with a matched group of failed reperfusion (n=42), patients reperfused after ≥6 passes (n=42) demonstrated a favorable shift in the overall modified Rankin Scale score distribution (adjusted odds ratio, 2.640 [95% CI, 1.073–6.686], P =0.037) and had higher rates of functional independence (36.8% vs 11.1%, P =0.004, adjusted odds ratio, 5.392 [95% CI, 1.185–24.530], P =0.029) at 90 day. Rates of parenchymal hematoma type-2 and 90-day mortality were comparable in the reperfused and nonreperfused groups. Conclusions: Achieving reperfusion despite multiple passes leads to improved outcomes compared with failed procedures. Arbitrary uncontrolled thresholds for a maximum number of passes to predict futile recanalization may lead to inappropriate early termination of procedures.

Introduction: The objective of our study was to identify the underlying causes of acute ischemic strokes (AIS) in patients prescribed direct oral anticoagulants (DOACs). Methods: This is a retrospective study of patients presenting to a large urban comprehensive stroke center from December 1, 2019 to May 31, 2020 who were prescribed DOACs at the index event. Two board-certified vascular neurologists independently reviewed medical charts including patient baseline characteristics, laboratory results, stroke mechanism, and patient-reported adherence. DOAC “failure” was defined as AIS occurring in the setting of self-reported adherence to the FDA-indicated dosing regimen. Results: Of 302 AIS patients admitted during the study period, 18 (6.0%) patients had AIS while on DOACs. Median age was 82 (IQR 72, 86) years, 9 (50%) were black and 14 (78%) were female. Eight patients (44%) presented within the IV Alteplase window and 5 (28%) patients presented with large vessel occlusion. At the time of stroke, 13 (72%) patients were on apixaban, 4 (22%) were on rivaroxaban and 1 (6%) on dabigatran. Atrial fibrillation (mean CHADS-VASc score 4.7) was the most common indication for DOAC (72%), followed by deep venous thrombosis (17%), and embolic stroke (11%). The most frequent causes of AIS on DOACs were underdosing in patients ≥ 80 years (n=6), known active malignancy (n=5), and non-adherence (n=4). The remaining cases included procedure-related discontinuation of the DOAC (n=1), symptomatic extracranial internal carotid artery stenosis (n=1), and unknown cause (n=1). Overall, the frequency of DOAC “failure” was 39%. In 6 (33%) patients, the initial anticoagulant was switched to an alternative anticoagulant at discharge. Conclusion: In this cohort of patients who had AIS while on DOAC therapy, 39% of cases were attributed to DOAC “failure” and primarily occurred in patients with active malignancy. Underdosing in the elderly and non-adherence remained the leading cause of AIS in patients on DOAC. Our findings highlight the importance of accurate dosing in the elderly and reinforcing adherence to medications. Further studies are needed to understand the role of DOACs in patients with active malignancy.

Introduction: The COVID-19 pandemic has wreaked havoc on the presentation, care and outcomes of patients with acute cerebrovascular and cardiovascular conditions. We sought to measure the national impact of COVID-19 on the care for acute ischemic stroke (AIS) and acute myocardial infarction (AMI). Methods: In this retrospective, observational study, we used the Premier Healthcare Database to evaluate the changes in the volume of care and hospital outcomes for AIS and AMI in relation to the pandemic. The pandemic months were defined from March 1, 2020- April 30, 2020 and compared to the same period in the year prior. Outcome measures were volumes of hospitalization and reperfusion treatment for AIS and AMI (including intravenous thrombolysis [IVT] and/or mechanical thrombectomy [MT] for AIS and percutaneous coronary interventions [PCI] for AMI) as well as in-hospital mortality, hospital length of stay (LOS) and hospitalization costs were compared across a 2-month period at the height of the pandemic versus the corresponding period in the prior year. Results: There were 95,453 AIS patients across 145 hospitals and 19,744 AMI patients across 126 hospitals. There was a significant nation-wide decline in the absolute number of hospitalizations for AIS (-38.94%;95%CI,-34.75% to -40.71%) and AMI (-38.90%;95%CI,-37.03% to -40.81%) as well as IVT (-30.32%;95%CI,-27.02% to -33.83%), MT (-23.54%;95%CI,-19.84% to -27.70%), and PCI (-35.05%;95%CI,-33.04% to -37.12%) during the first two months of the pandemic. This occurred across low-, mid-, and high-volume centers and in all geographic regions. Higher in-hospital mortality was observed in AIS patients (5.7% vs.4.2%, p=0.0037;OR 1.41,95%CI 1.1-1.8) but not AMI patients. A shift towards an increase in the proportion of admitted AIS and AMI patients receiving reperfusion therapies suggests a greater clinical severity among patients that were hospitalized for these conditions during the pandemic. A shorter length of stay (AIS: -17%, AMI: -20%), and decreased hospitalization costs (AIS: -12%, AMI: -19%) were observed. Conclusions: Our findings shed light on the combined health outcomes and economic impact the COVID-19 pandemic has had on acute stroke and cardiac emergency care.

Background: Access to endovascular therapy (EVT) should be equitable to all eligible ischemic stroke patients presenting within the treatment time window. In the Florida Stroke Registry (FSR) we sought to determine sex, race/ethnic, hospital and regional disparities in the delivery of EVT. Methods: From January 2010 to January 2020, a total of 99,088 ischemic stroke cases within 24 hours of symptom onset were enrolled. Multivariable logistic regression with generalized estimating equations evaluated independent predictors of EVT utilization. Results: A total of 7,812 patients received EVT (51.2% female, mean age 71.3 ± SD 14.6 years, 61.4% white, 17.4% black, 21.2% Hispanic). Compared to those not treated, EVT treated patients were more likely Hispanic (21.2% vs. 14.6%), arrived earlier to the hospital (median 120 min (IQR 292) vs. 170 min (IQR 446)), via EMS (94% vs. 66%), with more severe strokes (median NIHSS 15 (IQR11) vs. 5 (IQR 9)), to large hospitals (≥ 680 beds) (73.4% vs. 47.7%), in South Florida (50.8% vs. 38.2%). In multivariable analysis, female sex (OR 1.05, 95% CI 1-1.11), atrial fibrillation (OR 1.57, 95% CI 1.45-1.7), higher NIHSS (> 6) (OR 6.19, 95% CI 5.11-7.51) and presenting to a high-volume hospital (OR 3.47, 95% CI 2.25-5.36) positively predicted EVT utilization, whereas older age (>80 years) (OR 0.88, 95% CI 0.80-0.96), and black race (vs. white OR 0.87, 95% CI 0.76-1), were independently associated with lower use of EVT. Conclusions: In this large state-wide registry study, we found significant race-ethic and geographical disparities in delivery of EVT. Systems of care should address disparities in stroke treatment to improve access to EVT for all eligible stroke patients.

Introduction: Ticagrelor with aspirin has been recently shown to reduce the risk of stroke or death compared to aspirin alone in patients with high risk TIAs and mild strokes. However, this benefit is offset by increased risk of severe bleeding. We sought to evaluate the safety of ticagrelor in patients with moderate to severe ischemic stroke. Methods: This was a retrospective cohort study of adults discharged on ticagrelor after presenting with acute ischemic stroke and NIHSS > 5 from January 2016 to December 2019 at a large, urban, academic comprehensive stroke center. Patients were excluded if they underwent carotid or intracranial angioplasty and/or stenting, or carotid endarterectomy during admission. Baseline clinical characteristics, imaging, and outcomes were reviewed. Data was organized into continuous and categorical variables. Results: Sixty-one patients met inclusion and exclusion criteria. Median age was 61 (IQR, 52-68) years; 33 (54%) were men, and 33 (54%) were African American. Median NIHSS was 11 (IQR, 8-15). Fourteen (23%) patients received IV Alteplase and 35 (57%) patients underwent mechanical thrombectomy. Five (8%) patients received both IV Alteplase and mechanical thrombectomy. Median ticagrelor start date was hospital day 1 (IQR, 0-3). Large artery atherosclerosis was presumed etiology in 53 (87%) patients. No patients experienced neurologic worsening, recurrent stroke, sICH, or major bleeding during inpatient stay. Sixty (98%) patients were on aspirin and ticagrelor at discharge. Follow-up information was available for 53 (87%) patients for a median duration of 3 (IQR, 2-6) months. Following discharge, 3 (5%) patients experienced recurrent ischemic stroke despite being compliant. One (2%) patient experienced major bleeding—gastrointestinal hemorrhage requiring transfusion—two months after hospital discharge. Conclusions: This study highlights the potential expanding role for ticagrelor in secondary stroke prevention in patients with moderate to severe stroke. Early ticagrelor use did not result in sICH during inpatient stay—and only 1 major bleeding event on follow-up—in our cohort. While further research in this area is needed, these findings present an exciting opportunity for future prospective studies.

Background: Intraluminal carotid thrombi (ILT) are uncommon and present in only 0.4 to 1.5 % of patients with stroke. The endothelial inflammation and hypercoagulable state associated with COVID-19 (caused by SARS-CoV-2) poses a risk of arterial and venous thromboembolism. Herein, we present a series of stroke patients with ILT in the setting of a recent SARS-CoV-2 infection. Methods and Results: From July 9 th , 2020 to August 17 th , 2020, four patients were identified with acute ischemic stroke with varying thrombus burden but similar morphological patterns of ILT. All patients had evidence of recent SARS-CoV-2 infection as confirmed by PCR (n=3) or IgG antibody (n=1) in addition to clinical and radiological manifestations consistent with COVID-19. General characteristics and imaging of the patients are shown in table 1. Mean age was 55.3 ±7.7 years. Most patients (3) had at least two vascular risk factors and none had a central embolic source other than ILT. D-Dimer was profoundly elevated in 3 patients. ILT involved the cervical (3) and intracranial (1) internal carotid artery. COVID-19 symptoms ranged from mild transient respiratory symptoms to severe pneumonia requiring critical care support. None of the patients received acute reperfusion therapy either due to presentation beyond the therapeutic window or due to the absence of large vessel occlusion and/or unfavorable imaging profile. Conclusion: In this small case series of relatively young patients with acute ischemic stroke we found an association between COVID-19 infection and ILT. This suggests a distinctive pathophysiological pattern of stroke in COVID-19 patients. Further studies are necessary to elucidate the underlying pathophysiology and target specific treatment modalities.

Building an effective classifier for imbalanced data is a challenging task as most of classifier work on the assumption of balanced data. Therefore, several sampling methods have been devised to bridge this gap by re-sampling the imbalanced datasets. Although sampling methods are in abundance, there is no single method that is best suitable for all kinds of datasets and applications. Building classifiers for all the sampling methods and comparing the results using appropriate performance metrics is computationally inefficient. In this work, we propose a framework to find a relation between datasets and sampling methods via a set of meta-features that characterizes the distribution of data. Also, we take into account the effect of probability threshold on the choice of sampling methods. The main objective of this work is to develop an approach that aids the selection of one or more sampling methods together with a probability threshold to be used for building a suitable classifier for a given dataset. It is based on mapping functions learned between classifier performance and datasets after re-sampling. In this work, extensive experiments are performed to validate the framework using synthetic as well as KEEL benchmark datasets.

Background and Purpose The first pass effect (FPE) is a key predictor of clinical outcome after mechanical thrombectomy (MT) for large vessel occlusion stroke (LVOS). We aim to study the impact of initial angiographic appearance of middle cerebral artery (MCA) occlusion on FPE. Methods A prospectively maintained MT database from January 2018 to December 2019 was reviewed. Patients with M1 segment MCA occlusion were included. FPE was defined as complete reperfusion (mTICI2c-3) after a single pass thrombectomy. Occlusion site angiographic appearance was classified into two categories; regular, when the occlusion is smooth stump with abrupt cut off and irregular which included all other types of occlusion. The primary outcomes were the rates of FPE and functional independence (90-day mRS 0–2). Uni- and multivariable analyses were performed to identify predictors of FPE. A secondary analysis was the interaction between the first line MT technique and the shape of occlusion on the FPE. Results A total of 319 were included in the analysis, the mean age was 65.1 years, 168 (52.7%) were females. Median [IQR] baseline NIHSS score was 17 [12–21] and ASPECTS was 8 [7–9]. Patients with the regular shape of occlusion had a significantly higher frequency of smoking (28.5% vs. 18.5%, p=0.037). Cardioembolic strokes were more frequently presented with the regular shape of occlusion (53.1% vs. 43.9%). Conversely, intracranial atherosclerosis was significantly associated with irregular shape of occlusion (13.2% vs. 6%), p=0.04. Irregular occlusion shape had significantly lower median number of MT passes to achieve successful reperfusion (1 [1–2] vs. 2 [1–3], p=0.01 and higher rates of FPE 46% vs. 34.6%, p=0.04) and a trend to lower rates of parenchymal hematoma type 2 (2.6% vs. 6.9%, p=0.067). There was no significant interaction between first-line MT technique and the shape of occlusion on FPE. Functional independence and 90-day mortality were similar between the two groups. On multivariable analysis, the irregular shape of occlusion was an independent predictor of FPE (OR 1.756; 95% CI[1.049–2.940], p=0.03). Conclusions The angiographic shape of M1 occlusion is a predictor of FPE. Large prospective multicenter studies are warranted for a better understanding of whether the occlusion appearance guides the selection of the MT techniques and hence potentially increase the chance of FPE. Disclosures M. Mohammaden: None. D. Haussen: 2; C; Stryker, Vesalio. 4; C; Viz.AI. L. Pisani: None. A. Al-Bayati: None. S. Belagaje: None. N. Bhatt: None. M. Frankel: None. R. Nogueira: 2; C; Phenox, Anaconda, Genentech, Brainomix, Viz-AI

The Kalman filter and its variants have been developed for state estimation in semi-explicit, index-1 DAE systems in current literature. In this work, we develop a method for state estimation in non-linear fully-implicit, index-1 differential algebraic equation (DAEs) systems. In order to extend the Kalman filtering techniques for fully-implicit index-1 DAE systems, in the correction step we convert the fully-implicit DAE into a system of ordinary differential equations (ODEs). This is achieved by the index reduction of DAE using the method of successive differentiation of algebraic equations. This is a challenging problem as the fully-implicit DAE system does not contain explicit algebraic states unlike in the semi-explicit case. In this work, we propose a linear transformation of the mass matrix which enables us to find candidate algebraic states for the system. This transformation on the mass matrix is relatively simpler than constructing the transformation matrices in the Weierstrass-Kronecker canonical form. We illustrate our proposed method with two examples, a linear and a non-linear fully-implicit index-1 DAE system.