scholarly journals Epidemiological Surveillance of the Impact of the COVID-19 Pandemic on Stroke Care Using Artificial Intelligence

Stroke ◽  
2021 ◽  
Author(s):  
Raul G. Nogueira ◽  
Jason M. Davies ◽  
Rishi Gupta ◽  
Ameer E. Hassan ◽  
Thomas Devlin ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Acute Stroke ◽  
Systems Of Care ◽  
Ct Perfusion ◽  
Stroke Care ◽  
Large Vessel ◽  
Large Vessel Occlusion ◽  
Vessel Occlusion ◽  
Number Of Patients ◽  
The Impact

Background and Purpose: The degree to which the coronavirus disease 2019 (COVID-19) pandemic has affected systems of care, in particular, those for time-sensitive conditions such as stroke, remains poorly quantified. We sought to evaluate the impact of COVID-19 in the overall screening for acute stroke utilizing a commercial clinical artificial intelligence platform. Methods: Data were derived from the Viz Platform, an artificial intelligence application designed to optimize the workflow of patients with acute stroke. Neuroimaging data on suspected patients with stroke across 97 hospitals in 20 US states were collected in real time and retrospectively analyzed with the number of patients undergoing imaging screening serving as a surrogate for the amount of stroke care. The main outcome measures were the number of computed tomography (CT) angiography, CT perfusion, large vessel occlusions (defined according to the automated software detection), and severe strokes on CT perfusion (defined as those with hypoperfusion volumes >70 mL) normalized as number of patients per day per hospital. Data from the prepandemic (November 4, 2019 to February 29, 2020) and pandemic (March 1 to May 10, 2020) periods were compared at national and state levels. Correlations were made between the inter-period changes in imaging screening, stroke hospitalizations, and thrombectomy procedures using state-specific sampling. Results: A total of 23 223 patients were included. The incidence of large vessel occlusion on CT angiography and severe strokes on CT perfusion were 11.2% (n=2602) and 14.7% (n=1229/8328), respectively. There were significant declines in the overall number of CT angiographies (−22.8%; 1.39–1.07 patients/day per hospital, P <0.001) and CT perfusion (−26.1%; 0.50–0.37 patients/day per hospital, P <0.001) as well as in the incidence of large vessel occlusion (−17.1%; 0.15–0.13 patients/day per hospital, P <0.001) and severe strokes on CT perfusion (−16.7%; 0.12–0.10 patients/day per hospital, P <0.005). The sampled cohort showed similar declines in the rates of large vessel occlusions versus thrombectomy (18.8% versus 19.5%, P =0.9) and comprehensive stroke center hospitalizations (18.8% versus 11.0%, P =0.4). Conclusions: A significant decline in stroke imaging screening has occurred during the COVID-19 pandemic. This analysis underscores the broader application of artificial intelligence neuroimaging platforms for the real-time monitoring of stroke systems of care.

Abstract 42: Epidemiological Surveillance of the Impact of the Covid-19 Pandemic on Stroke Care Using Artificial Intelligence

Stroke ◽  
2021 ◽  
Vol 52 (Suppl_1) ◽  
Author(s):  
Raul Nogueira ◽  
Jason Davies ◽  
Rishi Gupta ◽  
Ameer E Hassan ◽  
Thomas G Devlin ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Acute Stroke ◽  
Real Time ◽  
Systems Of Care ◽  
Stroke Care ◽  
Epidemiological Surveillance ◽  
Day Hospital ◽  
Stroke Patients ◽  
Number Of Patients ◽  
The Impact

Background: The degree to which the COVID-19 pandemic has affected systems of care, in particular those for time-sensitive conditions such as stroke, remains poorly quantified. We sought to evaluate the impact of COVID-19 in the overall screening for acute stroke utilizing a commercial clinical artificial intelligence (AI) platform. Methods: Data were derived from the Viz Platform, an AI application designed to optimize the workflow of acute stroke patients. Neuroimaging data on suspected stroke patients across 97 hospitals in 20 US states were collected in real-time and retrospectively analyzed with the number of patients undergoing imaging screening serving as a surrogate for the amount of stroke care. The main outcome measures were the number of CTA, CTP, Large vessel occlusions (LVOs) (defined according to the automated software detection), and severe strokes on CTP (defined as those with hypoperfusion volumes>70mL) normalized as number of patients per day per hospital. Data from the pre-pandemic (November 4, 2019 to February 29, 2020) and pandemic (March 1 to May 10, 2020) periods were compared at national and state levels. Correlations were made between the inter-period changes in imaging screening, stroke hospitalizations, and thrombectomy procedures using state-specific sampling. Results: A total of 23,223 patients were included. The incidence of LVO on CTA and severe strokes on CTP were 11.2%(n=2,602) and 14.7%(n=1,229/8,328), respectively. There were significant declines in the overall number of CTAs (-22.8%;1.39 to 1.07 patients/day/hospital,p<0.001) and CTPs (-26.1%;0.50 to 0.37 patients/day/hospital,p<0.001) as well as in the incidence of LVO (-17.1%;0.15 to 0.13 patients/day/hospital,p<0.001) and severe strokes on CTP (-16.7%;0.12 to 0.10 patients/day/hospital, p<0.005). The sampled cohort showed similar declines in the rates of LVOs versus thrombectomy (18.8%vs.19.5%, p=0.9) and CSC hospitalizations (18.8%vs.11.0%, p=0.4). Conclusions: A significant decline in stroke imaging screening has occurred during the COVID-19 pandemic. This underscores the broader application of AI neuroimaging platforms for the real-time monitoring of stroke systems of care.

Abstract TP77: Impact of Time and Aspects on Clinical Core Mismatch in Large Vessel Occlusion Strokes

Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Shashvat Desai ◽  
Bradley J Molyneaux ◽  
Marcelo Rocha ◽  
Matthew Starr ◽  
Tudor G Jovin ◽  
...  
Keyword(s):  
Patient Selection ◽  
Time Window ◽  
Ct Perfusion ◽  
Large Vessel ◽  
Large Vessel Occlusion ◽  
Vessel Occlusion ◽  
Ct Head ◽  
Ischemic Core ◽  
The Impact ◽  
Core Volume

Introduction: Patient selection for endovascular thrombectomy (EVT) for anterior circulation large vessel occlusion (LVO) strokes in the 6-24-hour time window is dependent on delineating clinical core mismatch (CCM) as defined by DAWN trial criteria. In contrast, patient selection in the early window (0-6 hours) can be performed using ASPECTS on CT head. We aim to determine the prevalence of DAWN-CCM in LVO strokes and the impact of time and ASPECTS. Methods: Retrospective analysis of large vessel occlusion [internal carotid and middle cerebral artery-M1] strokes at a CSC. Consecutive patients who underwent CT perfusion or MRI within 120 minutes of CT head were included in the study (treated and untreated). Ischemic core volume was assessed using RAPID [IschemaView] and ASPECTS using automated ASPECTS [Brainomix]. CCM was defined using DAWN trial criteria [DAWN-CCM: NIHSS ≥10 and core <31 ml, NIHSS ≥20 and core <51 ml]. Results: A total of 116 patients were included. Mean age was 71 ±14 and 62% were females. Mean ischemic core volume and median ASPECTS were 46 ±65 ml and 8 (6-9), respectively. In patients with NIHSS score ≥10 (98), 57% had DAWN-CCM in the 0-24-hour window. Proportion of patients with DAWN-CCM in 6-24-hour window was 70% (6-12 hours), 50% (12-18 hours), and 50% (18-24 hours) [p=0.35]. Proportion of patients with DAWN-CCM by ASPECTS group was 88% (ASPECTS 9-10), 64% (ASPECTS 6-8) and 13% (ASPECTS 0-5) [p=<0.01] (Figure 1). Probability of DAWN-CCM declines by 7% for every 2 hours increase in TLKW to imaging, and by 13% for every 1-point decrease in ASPECTS. Conclusion: Approximately 57% of LVO strokes have clinical core mismatch. LVO strokes with DAWN-CCM decline with increasing time and decreasing ASPECTS. ASPECTS alone may be sufficient to identify patients with DAWN-CCM in a resource limited setting and avoid time consuming advanced imaging.

Abstract TMP8: Predictors of Large Vessel Occlusion in Kaiser Stroke EXPRESS Program

Stroke ◽  
2017 ◽  
Vol 48 (suppl_1) ◽  
Author(s):  
Jeffrey G Klingman ◽  
Anne C Kim ◽  
Meghan Hatfield ◽  
Benjamin Wilson ◽  
Lauren Klingman ◽  
...  
Keyword(s):  
Acute Stroke ◽  
Patient Characteristics ◽  
Stroke Care ◽  
Large Vessel ◽  
Large Vessel Occlusion ◽  
Acute Stroke Care ◽  
Kaiser Permanente ◽  
Vessel Occlusion ◽  
Stroke Treatment ◽  
Improved Outcomes

Background: In 2015, trials showed that rapid endovascular stroke treatment (EST) of qualified patients with large vessel occlusion (LVO) resulted in improved outcomes over treatment with IV tPA alone. In 2015, Kaiser Permanente Northern California (KPNC) redesigned its acute stroke care work flow for its 21 stroke centers, which included expedited IV t-pa treatment, rapid CTA investigation, expedited transfer of appropriate patients for EST. We assessed for predictors of LVO post-implementation. Methods: The KPNC Stroke EXPRESS program was live in all centers by January 2016. Using clinical data for 1/1/16 - 7/10/16, we evaluated the frequency and locations of LVO, and patient characteristics of those with LVO. Multivariate logistic regression was used to examine whether age, gender, race, or an NIHSS ≥ 8 are predictors of LVO. Results: There were 2,204 tele-stroke alert cases from the ED. Among 993 (39.3%) that proceeded as likely acute stroke, 812 (81.8%) were evaluated with CTA. Out of those who had a CTA, 152 (18.7%) were found to have LVO as followed: 27 (17.8%) ICA, 87 (57.2%) M1, 24 (15.8%) M2, 6 (4.0%) basilar, 5 (3.3%) PCA, and 3 (2.0%) vertebral. Of those with LVO, 97 (63.8%) were treated with EST. Patients with LVO had a higher median NIHSS (15 vs. 5 in those without LVO). Neglect (27% vs. 7%) and gaze deviation (16% vs. 1%) were more likely to be seen among those with LVO and treated with EST compared to those without LVO. In multivariate analysis, age (OR=1.02, 95% CI 1.00 - 1.03, p=0.01) and NIHSS ≥8 (OR = 4.99, 95% CI 3.32- 7.49, p < 0.001) were associated with LVO. PPV for NIHSS ≥8 was 75.7%. Conclusions: In our large multi-ethnic population of acute stroke patients, a relatively small percentage (19%) was found to have LVO and only a subset qualified for EST. Predictors of LVO included NIHSS ≥8, increasing age, and presence of neglect and gaze preference. Given the low numbers of patients brought in for acute stroke treatment who ended up with a LVO requiring EST, further research is needed to assess a given system’s ability to rapidly evaluate and transfer as appropriate for EST rather than paramedic based diversion.

Abstract 60: Impact of Monthly Testing and Tutorials on Accuracy and Confidence of Telestroke Nurse Recognition of Anterior Circulation Large Vessel Occlusion CT Perfusion Patterns

Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Emily K Snider ◽  
Brenda Chang ◽  
Anna Maria Helms ◽  
Laura Williams ◽  
Sam Singh ◽  
...  
Keyword(s):  
Good Accuracy ◽  
Internal Carotid ◽  
Ct Perfusion ◽  
Large Vessel ◽  
Large Vessel Occlusion ◽  
Anterior Circulation ◽  
Vessel Occlusion ◽  
Confidence Levels ◽  
Lower Confidence ◽  
The Impact

Introduction: A distinguishing feature of our Stroke Network is Telestroke (TS) nurses (RNs) who remotely facilitate TS evaluations. To facilitate expeditious transfer the TS RNs, need to immediately identify CT perfusion (CTP) studies demonstrating patterns consistent with internal carotid artery (ICA), middle cerebral artery (M1), and first branch of the M1 (M2) large vessel occlusion (LVO) acute ischemic strokes (AIS). Methods: We developed a 6-month series of tutorials and tests (12 CTP scans/month) for 16 TS RNs to recognize CTP patterns consistent with ICA, M1 or M2 LVO AIS. We simultaneously conducted a prospective cohort study of these nurses on the impact of these tutorials and tests. Results: TS nurses demonstrated good accuracy in detecting ICA, M1 or M2 LVO presence during the first three months of teaching (83-94% accurate).This improved to excellent during the last three months (99-100%), during which the likelihood of correctly identifying the presence of any one of these LVOs was greater than in the first three months (OR 1.99, 95% CI: 1.83-2.17, p<0.001). The probability of correctly identifying ICA or M1 occlusions was much higher than correctly identifying all other LVOs (OR 68, 95% CI: 45-102, p<0.001). The likelihood of being at a higher confidence level compared to lower confidence levels in identifying any LVOs as being ICA or M1 was higher than in identifying other LVOs (OR 2.14, 95% CI: 1.6-2.8, p<0.001), but over time confidence for determining LVO presence did not differ significantly after controlling for subject variation. Conclusion: A series of structured tutorials significantly increased the odds of TS nurses correctly identifying anterior circulation LVOs, with the benefit of these tutorials and test reviews peaking and plateauing at four months. Participating in TS nurse tutorials was associated with high odds and confidence for correctly identifying LVOs as being ICA or M1.

scholarly journals Performance of Automated Attenuation Measurements at Identifying Large Vessel Occlusion Stroke on CT Angiography

2020 ◽  
Author(s):  
Paul Reidler ◽  
Lena Stueckelschweiger ◽  
Daniel Puhr-Westerheide ◽  
Katharina Feil ◽  
Lars Kellert ◽  
...  
Keyword(s):  
Visual Analysis ◽  
Ct Perfusion ◽  
Characteristic Curve ◽  
Large Vessel ◽  
Youden Index ◽  
Control Group ◽  
Large Vessel Occlusion ◽  
Stroke Patients ◽  
Vessel Occlusion ◽  
Number Of Patients

Abstract Purpose Computed tomography angiography (CTA) is routinely used to detect large-vessel occlusion (LVO) in patients with suspected acute ischemic stroke; however, visual analysis is time consuming and prone to error. To evaluate solutions to support imaging triage, we tested performance of automated analysis of CTA source images (CTASI) at identifying patients with LVO. Methods Stroke patients with LVO were selected from a prospectively acquired cohort. A control group was selected from consecutive patients with clinically suspected stroke without signs of ischemia on CT perfusion (CTP) or infarct on follow-up. Software-based automated segmentation and Hounsfield unit (HU) measurements were performed on CTASI for all regions of the Alberta Stroke Program Early CT score (ASPECTS). We derived different parameters from raw measurements and analyzed their performance to identify patients with LVO using receiver operating characteristic curve analysis. Results The retrospective analysis included 145 patients, 79 patients with LVO stroke and 66 patients without stroke. The parameters hemispheric asymmetry ratio (AR), ratio between highest and lowest regional AR and M2-territory AR produced area under the curve (AUC) values from 0.95–0.97 (all p < 0.001) for detecting presence of LVO in the total population. Resulting sensitivity (sens)/specificity (spec) defined by the Youden index were 0.87/0.97–0.99. Maximum sens/spec defined by the specificity threshold ≥0.70 were 0.91–0.96/0.77–0.83. Performance in a small number of patients with isolated M2 occlusion was lower (AUC: 0.72–0.85). Conclusion Automated attenuation measurements on CTASI identify proximal LVO stroke patients with high sensitivity and specificity. This technique can aid in accurate and timely patient selection for thrombectomy, especially in primary stroke centers without CTP capacity.

scholarly journals Stroke Intervention: Geographic Disparities in Acute Stroke Care and the Role of Interventional Cardiology

2020 ◽  
Keyword(s):  
Acute Stroke ◽  
Case Series ◽  
Intravenous Thrombolysis ◽  
Interventional Cardiology ◽  
Stroke Care ◽  
Large Vessel ◽  
Large Vessel Occlusion ◽  
Acute Stroke Care ◽  
Vessel Occlusion ◽  
Geographic Disparities

Stroke is the second-leading cause of death and a major cause of disability worldwide. The majority of strokes are ischaemic, and effective therapy to achieve reperfusion includes intravenous thrombolysis and, for proximal large vessel occlusion strokes, endovascular mechanical thrombectomy (MT). There has been a paradigm shift in acute stroke care, driven by a series of randomised controlled trials demonstrating that timely reperfusion with MT results in superior outcomes compared to intravenous thrombolysis in patients with large vessel occlusion strokes. There are significant geographic disparities in delivering acute stroke care because of the maldistribution of neurointerventional specialists. There are now several case series demonstrating the feasibility and safety of first medical contact MT by carotid stent-capable interventional cardiologists and noninvasive neurologists working on stroke teams, which is a solution to the uneven distribution of neurointerventionalists and allows stroke interventions to be delivered in local communities.

scholarly journals CT perfusion: stroke, seizure or both?

2021 ◽  
Vol 14 (12) ◽  
pp. e245723
Author(s):  
Elizebath Davies ◽  
Fathalla Elnagi ◽  
Thomas Smith
Keyword(s):  
Acute Stroke ◽  
Small Vessel Disease ◽  
Ct Perfusion ◽  
Stroke Care ◽  
Large Vessel ◽  
Arterial Calcification ◽  
Acute Stroke Care ◽  
Vessel Occlusion ◽  
History Of ◽  
Stroke Mimic

An 88-year-old male with a history of hypertension, ischaemic heart disease and Bell’s palsy presented with symptoms and signs of an acute ischaemic stroke. National Institutes of Health Stroke Scale (NIHSS) was 19 at presentation, indicative of potential large vessel occlusion. The initial CT scan revealed evidence of small vessel disease and arterial calcification. As there were no contraindications, he received thrombolytic treatment. CT angiography and CT perfusion imaging were performed in preparation for possible thrombectomy. There was no evidence of a large vessel thrombus, and changes on CT perfusion were suggestive of seizure activity, with relative hyperperfusion on the cerebral hemisphere of interest. Post thrombolysis, his NIHSS was 5. An MR scan revealed evidence of bilateral thalamic infarcts. After a period of rehabilitation, he was discharged home and independently mobile but with cognitive impairment.Acute stroke care increasingly uses multimodal imaging to confirm the clinical diagnosis and help optimise initial emergency management. Such imaging is useful in determining whether the presentation is a vascular event or stroke mimic. Moreover, seizures complicate and mimic acute strokes, which can lead to therapeutic uncertainty. This case highlights the increasingly sophisticated investigation of patients presenting with suspected acute stroke, with the attendant need for accurate interpretation by experienced clinicians.

Abstract WP334: Stroke Case Volume After Implementation of the 24 Hour Extended Intra-Arterial Thrombectomy Window

Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Jessica Glenn ◽  
Sharon Heaton ◽  
Ciarán Powers ◽  
Shahid Nimjee ◽  
Patrick Youssef ◽  
...  
Keyword(s):  
Systems Of Care ◽  
Ct Perfusion ◽  
Stroke Patients ◽  
Case Volume ◽  
Anterior Circulation ◽  
Vessel Occlusion ◽  
Number Of Patients ◽  
Comprehensive Stroke Center ◽  
The Impact ◽  
Treatment Window

Introduction: The results of the DAWN and DEFUSE 3 trials demonstrated improved functional outcome with mechanical thrombectomy (MT) in stroke patients who present with anterior circulation large vessel occlusion (LVO) up to a 24 hours after last known well. The AHA guidelines were updated to include this recommendation in 2018, but the impact of the increase in number of patients treated in clinical practice remains unknown. Methods: We reviewed ischemic stroke MT patients admitted to a comprehensive stroke center (DEFUSE 3 trial participating site), from July 2015- July 2019. Our center instituted a 24 hour IAT window protocol in March 2018 (Figure 1). Eligibility for MT in the 24 hour window followed the DEFUSE 3 criteria including the presence of LVO, National Institute of Health Stroke Scale (NIHSS) > 6, and favorable CT perfusion imaging. Results: During this 4-year period, 290 stroke patients with LVO received MT at our institution. MT volumes increased in each subsequent year in this study as follows: July 2015 (44), July 2016 (58), July 2017 (73) and July 2018 (115). An increase in MT volume was observed after the expansion of the treatment window to 24 hours (Figure 2). In the period prior to the expanded 24 hour treatment window, MT volumes were 25-33 per 6 month period. After the 24 hour treatment window was started, MT volumes were 58-60 per 6 month period. Conclusions: Expanding the MT window to 24 hours has nearly doubled the MT case volume at our single center. This dramatic increase in IAT clinical utilization observed with expansion to the 24 hour window has implications for systems of care and resource optimization of the stroke patient.

scholarly journals Diagnostic value of transcranial ultrasonography for selecting subjects with large vessel occlusion: a systematic review

2019 ◽  
Vol 11 (1) ◽  
Author(s):  
Daria Antipova ◽  
Leila Eadie ◽  
Ashish Stephen Macaden ◽  
Philip Wilson
Keyword(s):  
Systematic Review ◽  
Diagnostic Accuracy ◽  
Acute Stroke ◽  
Clinical Assessment ◽  
Assessment Tools ◽  
Large Vessel ◽  
Cochrane Library ◽  
Direct Transfer ◽  
Large Vessel Occlusion ◽  
Vessel Occlusion

Abstract Introduction A number of pre-hospital clinical assessment tools have been developed to triage subjects with acute stroke due to large vessel occlusion (LVO) to a specialised endovascular centre, but their false negative rates remain high leading to inappropriate and costly emergency transfers. Transcranial ultrasonography may represent a valuable pre-hospital tool for selecting patients with LVO who could benefit from rapid transfer to a dedicated centre. Methods Diagnostic accuracy of transcranial ultrasonography in acute stroke was subjected to systematic review. Medline, Embase, PubMed, Scopus, and The Cochrane Library were searched. Published articles reporting diagnostic accuracy of transcranial ultrasonography in comparison to a reference imaging method were selected. Studies reporting estimates of diagnostic accuracy were included in the meta-analysis. Results Twenty-seven published articles were selected for the systematic review. Transcranial Doppler findings, such as absent or diminished blood flow signal in a major cerebral artery and asymmetry index ≥ 21% were shown to be suggestive of LVO. It demonstrated sensitivity ranging from 68 to 100% and specificity of 78–99% for detecting acute steno-occlusive lesions. Area under the receiver operating characteristics curve was 0.91. Transcranial ultrasonography can also detect haemorrhagic foci, however, its application is largely restricted by lesion location. Conclusions Transcranial ultrasonography might potentially be used for the selection of subjects with acute LVO, to help streamline patient care and allow direct transfer to specialised endovascular centres. It can also assist in detecting haemorrhagic lesions in some cases, however, its applicability here is largely restricted. Additional research should optimize the scanning technique. Further work is required to demonstrate whether this diagnostic approach, possibly combined with clinical assessment, could be used at the pre-hospital stage to justify direct transfer to a regional thrombectomy centre in suitable cases.

Sign in / Sign up

Export Citation Format

Share Document