Project description:Background:The detection of large vessel occlusion (LVO) plays a critical role in the diagnosis and treatment of acute ischemic stroke (AIS). Identifying LVO in the pre-hospital setting or early stage of hospitalization would increase the patients' chance of receiving appropriate reperfusion therapy and thereby improve neurological recovery. Methods:To enable rapid identification of LVO, we established an automated evaluation system based on all recorded AIS patients in Hong Kong Hospital Authority's hospitals in 2016. The 300 study samples were randomly selected based on a disproportionate sampling plan within the integrated electronic health record system, and then separated into a group of 200 patients for model training, and another group of 100 patients for model performance evaluation. The evaluation system contained three hierarchical models based on patients' demographic data, clinical data and non-contrast CT (NCCT) scans. The first two levels of modeling utilized structured demographic and clinical data, while the third level involved additional NCCT imaging features obtained from deep learning model. All three levels' modeling adopted multiple machine learning techniques, including logistic regression, random forest, support vector machine (SVM), and eXtreme Gradient Boosting (XGboost). The optimal cut-off for the likelihood of LVO was determined by the maximal Youden index based on 10-fold cross-validation. Comparisons of performance on the testing group were made between these techniques. Results:Among the 300 patients, there were 160 women and 140 men aged from 27 to 104 years (mean 76.0 with standard deviation 13.4). LVO was present in 130 (43.3%) patients. Together with clinical and imaging features, the XGBoost model at the third level of evaluation achieved the best model performance on testing group. The Youden index, accuracy, sensitivity, specificity, F1 score, and area under the curve (AUC) were 0.638, 0.800, 0.953, 0.684, 0.804, and 0.847, respectively. Conclusion:To the best of our knowledge, this is the first study combining both structured clinical data with non-structured NCCT imaging data for the diagnosis of LVO in the acute setting, with superior performance compared to previously reported approaches. Our system is capable of automatically providing preliminary evaluations at different pre-hospital stages for potential AIS patients.

Project description:Background and Purpose- Accurate prediction of acute ischemic stroke (AIS) caused by anterior large vessel occlusion (LVO) that is amendable to mechanical thrombectomy remains a challenge. We developed and validated a prediction model for anterior circulation LVO stroke using past medical history elements present on admission and neurological examination. Methods- We retrospectively reviewed AIS patients admitted between 2009 and 2017 to 3 hospitals within a large healthcare system in the United States. Patients with occlusions of the internal carotid artery or M1 or M2 segments of the middle cerebral artery were randomly split into 2/3 derivation and 1/3 validation cohorts for development of an anterior circulation LVO prediction model and score that was further curtailed for potential use in the prehospital setting. Results- A total of 1654 AIS were reviewed, including 248 (15%) with proximal anterior circulation LVO AIS. In the derivation cohort, National Institutes of Health Stroke Scale score at the time of cerebrovascular imaging, current smoking status, type 2 diabetes mellitus, extracranial carotid, and intracranial atherosclerotic stenosis was significantly associated with anterior circulation LVO stroke. The prehospital score was curtailed to National Institutes of Health Stroke Scale score, current smoking status, and type 2 diabetes mellitus. The areas under the curve for the prediction model, prehospital score, and National Institutes of Health Stroke Scale score alone were 0.796, 0.757, and 0.725 for the derivation cohort and 0.770, 0.689, and 0.665 for the validation cohort, respectively. The Youden index J was 0.46 for a score of >6 with 84.7% sensitivity and 62.0% specificity for the prediction model. Conclusions- Previously reported LVO stroke prediction scores focus solely on elements of the neurological examination. In addition to stroke severity, smoking, diabetes mellitus, extracranial carotid, and intracranial atherosclerotic stenosis were associated with anterior circulation LVO AIS. Although atherosclerotic stenosis may not be known until imaging is obtained, smoking and diabetes mellitus history can be readily obtained in the field and represent important elements of the prehospital score supplementing National Institutes of Health Stroke Scale score.

Project description:BackgroundNeuroinflammation plays an important role in the pathogenesis of acute ischemic stroke (AIS) and peripheral leukocyte counts have proved to be independent predictors of stroke severity and outcomes. Clinical significance of large vessel occlusion (LVO) in AIS is increasing, as these patients are potential candidates for endovascular thrombectomy and likely to have worse outcomes if not treated urgently. The aim of our study was to assess the relationship between on admission leukocyte counts and the presence of LVO in the early phase of AIS.MethodsWe have conducted a cross-sectional, observational study based on a registry of consecutive AIS patients admitted up to 4.5 h after stroke onset. Blood samples were taken at admission and leukocyte counts were measured immediately. The presence of LVO was verified based on the computed tomography angiography scan on admission.ResultsTotal white blood cell (WBC) and neutrophil counts were significantly higher in patients with LVO than those without LVO (P < 0.001 respectively). After adjustment for potential confounders total WBC counts (adjusted OR: 1.405 per 1 × 109/L increase, 95% CI: 1.209 to 1.632) and neutrophil counts (adjusted OR: 1.344 per 1 × 109/L increase, 95% CI: 1.155 to 1.564) were found to have the strongest associations with the presence of LVO. Total WBC and neutrophil counts had moderate ability to discriminate an LVO in AIS (AUC: 0.667 and 0.655 respectively). No differences were recorded in leukocyte counts according to the size of the occluded vessel and the status of collateral circulation in the anterior vascular territory. However, total WBC and neutrophil counts tended to be higher in patients with LVO in the posterior circulation (p = 0.005 and 0.010 respectively).ConclusionHigher admission total WBC and neutrophil counts are strongly associated with the presence of LVO and has moderate ability to discriminate an LVO in AIS. Detailed evaluation of stroke-evoked inflammatory mechanisms and changes according to the presence of LVO demands further investigation.

Project description:BackgroundThe management of acute ischemic stroke (AIS) is time-sensitive, yet prehospital delays remain prevalent. The application of large language models (LLMs) for medical text analysis may play a potential role in clinical decision support. We assess the performance of LLMs on prehospital AIS and large vessel occlusion (LVO) stroke screening.MethodsThis retrospective study sourced cases from the electronic medical record database of the emergency department (ED) at Maoming People's Hospital, encompassing patients who presented to the ED between June and November 2023. We evaluate the diagnostic accuracy of GPT-3.5 and GPT-4 for the detection of AIS and LVO stroke by comparing the sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and positive likelihood ratio and AUC of both LLMs. The neurological reasoning of LLMs was rated on a five-point Likert scale for factual correctness and the occurrence of errors.ResultOn 400 records from 400 patients (mean age, 70.0 years ± 12.5 [SD]; 273 male), GPT-4 outperformed GPT-3.5 in AIS screening (AUC 0.75 (0.65-0.84) vs 0.59 (0.50-0.69), P = 0.015) and LVO identification (AUC 0.71 (0.65-0.77) vs 0.60 (0.53-0.66), P < 0.001). GPT-4 achieved higher accuracy than GPT-3.5 in screening of AIS (89.3% [95% CI: 85.8, 91.9] vs 86.5% [95% CI: 82.8, 89.5]) and LVO stroke identification (67.0% [95% CI: 62.3%, 71.4%] vs 47.3% [95% CI: 42.4%, 52.2%]). In neurological reasoning, GPT-4 had higher Likert scale scores for factual correctness (4.24 vs 3.62), with a lower rate of error (6.8% vs 24.8%) than GPT-3.5 (all P < 0.001).ConclusionsThe result demonstrates that LLMs possess diagnostic capability in the prehospital identification of ischemic stroke, with the ability to exhibit neurologically informed reasoning processes. Notably, GPT-4 outperforms GPT-3.5 in the recognition of AIS and LVO stroke, achieving results comparable to prehospital scales. LLMs are supposed to become a promising supportive decision-making tool for EMS practitioners in screening prehospital stroke.

Project description:BACKGROUND:Acute ischemic stroke (AIS) due to large vessel occlusion (LVO) is a devastating cerebrovascular disorder, which could benefit from collateral circulation. Proteins associated with acute LVO pathogenesis and endothelial function may appear in blood samples of AIS patients due to LVO, thus permitting development of blood-based biomarkers for its diagnosis and prognosis. METHODS:This study is a single-center, retrospective, observational case-control trial. Consecutive patients who presented at the Department of Neurology of Tongji Hospital were recruited from July 2016 to April 2018. In the discovery phase, a proteomic approach with iTRAQ-based LC-MS/MS was used to investigate the altered proteomic pattern in plasma from patients with AIS due to LVO. In the validation study, Western blots was used to identify biomarkers associated with stroke diagnosis as well as their prognostic value associated with different collateral statuses. RESULTS:For this exploratory study, the proteomic analysis of plasma from 40 patients with AIS due to LVO and 20 healthy controls revealed seven differentially expressed proteins with a 1.2/0.83-fold or greater difference between groups. The four elevated proteins, PPBP (1.58 ± 0.78 vs 0.98 ± 0.37; P < 0.001), THBS1 (1.13 ± 0.88 vs 0.43 ± 0.26; P < 0.001), LYVE1 (1.61 ± 0.55 vs 0.97 ± 0.50; P < 0.001), and IGF2 (1.19 ± 0.42 vs 0.86 ± 0.24; P < 0.001), were verified by Western blots analysis in an independent cohort including 33 patients and 33 controls. A strong interaction was observed between the four-protein panel and the diagnosis of AIS due to LVO (AUC 0.947; P < 0.001). Furthermore, IGF2, LYVE1, and THBS1 were closely associated with collateral status (IGF2 0.115, 95% CI 0.016-0.841, P = 0.033; LYVE1 0.183, 95% CI 0.036-0.918, P = 0.039; THBS1 4.257, 95% CI 1.273-14.228, P = 0.019), and proved to be independent predictors of good outcome (IGF2 0.115, 95% CI 0.015-0.866, P = 0.036; LYVE1 0.028, 95% CI 0.002-0.334, P = 0.005; THBS1 3.294, 95% CI 1.158-9.372, P = 0.025) at a 3-month follow-up. CONCLUSIONS:The identified 4-biomarker panel could provide diagnostic aid to the existing imaging modalities for AIS due to LVO, and the prognostic value of IGF2, LYVE1, and THBS1 was proved in predicting functional outcomes related to collateral status. Trial registration ClinicalTrials.gov NCT03122002. Retrospectively registered April 20, 2017. URL of trial registry record: https://www.clinicaltrials.gov/ct2/show/NCT03122002?term=NCT+03122002&rank=1.

Project description:The early detection of large-vessel occlusion (LVO) strokes is increasingly important as these patients are potential candidates for endovascular therapy, the availability of which is limited. Prehospital LVO detection scales mainly contain symptom variables only; however, recent studies revealed that other types of variables could be useful as well. Our aim was to comprehensively assess the predictive ability of several clinical variables for LVO prediction and to develop an optimal combination of them using machine learning tools. We have retrospectively analysed data from a prospectively collected multi-centre stroke registry. Data on 41 variables were collected and divided into four groups (baseline vital parameters/demographic data, medical history, laboratory values, and symptoms). Following the univariate analysis, the LASSO method was used for feature selection to select an optimal combination of variables, and various machine learning methods (random forest (RF), logistic regression (LR), elastic net method (ENM), and simple neural network (SNN)) were applied to optimize the performance of the model. A total of 526 patients were included. Several neurological symptoms were more common and more severe in the group of LVO patients. Atrial fibrillation (AF) was more common, and serum white blood cell (WBC) counts were higher in the LVO group, while systolic blood pressure (SBP) was lower among LVO patients. Using the LASSO method, nine variables were selected for modelling (six symptom variables, AF, chronic heart failure, and WBC count). When applying machine learning methods and 10-fold cross validation using the selected variables, all models proved to have an AUC between 0.736 (RF) and 0.775 (LR), similar to the performance of National Institutes of Health Stroke Scale (AUC: 0.790). Our study highlights that, although certain neurological symptoms have the best ability to predict an LVO, other variables (such as AF and CHF in medical history and white blood cell counts) should also be included in multivariate models to optimize their efficiency.

Dataset Information

Treating acute large vessel occlusion stroke: to bridge or not to bridge?

Publications

Treating acute large vessel occlusion stroke: to bridge or not to bridge?

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets