e-ASPECTS software improves interobserver agreement and accuracy of interpretation of aspects score

Brinjikji, Waleed; Abbasi, Mehdi; Arnold, Catherine; Benson, John C.; Braksick, Sherri A.; Campeau, Norbert G.; Carr, Carrie M.; Cogswell, Petrice M; Klaas, James P.; Liebo, Greta B.; Little, Jason T.; Luetmer, Patrick H.; Messina, Steven A.; Nagelschneider, Alex A.; Schwartz, Kara M.; Wood, Christopher P.; Nasr, Deena M.; Kallmes, David F.

doi:10.1177/15910199211011861

Cited by 22 publications

(18 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Quantitative imaging feature extraction was performed using e-Stroke software (Brainomix, Oxford, UK) for automated calculation of Alberta Stroke Program Early CT Score (ASPECTS; e-ASPECTS) and estimated acute infarct volumes on non-contrast CT ( 14 – 17 ). e-ASPECTS uses ML classification to distinguish and segment regions that contain signs consistent with the acute ischemic change in order to output both (total and per ASPECTS region) and total e-ASPECTS volumes.…”

Section: Methodsmentioning

confidence: 99%

Interpretable Machine Learning Modeling for Ischemic Stroke Outcome Prediction

et al. 2022

Self Cite

View full text Add to dashboard Cite

Background and PurposeMechanical thrombectomy greatly improves stroke outcomes. Nonetheless, some patients fall short of full recovery despite good reperfusion. The purpose of this study was to develop machine learning (ML) models for the pre-interventional prediction of functional outcome at 3 months of thrombectomy in acute ischemic stroke (AIS), using clinical and auto-extractable radiological information consistently available upon first emergency evaluation.Materials and MethodsA two-center retrospective cohort of 293 patients with AIS who underwent thrombectomy was analyzed. ML models were developed to predict dichotomized modified Rankin score at 90 days (mRS-90) using clinical and imaging features, both separately and combined. Conventional and experimental imaging biomarkers were quantified using automated image-processing software from non-contract computed tomography (CT) and computed tomography angiography (CTA). Shapley Additive Explanation (SHAP) was applied for model interpretability and predictor importance analysis of the optimal model.ResultsMerging clinical and imaging features returned the best results for mRS-90 prediction. The best performing classifier was Extreme Gradient Boosting (XGB) with an area under the receiver operating characteristic curve (AUC) = 84% using selected features. The most important classifying features were age, baseline National Institutes of Health Stroke Scale (NIHSS), occlusion side, degree of brain atrophy [primarily represented by cortical cerebrospinal fluid (CSF) volume and lateral ventricle volume], early ischemic core [primarily represented by e-Alberta Stroke Program Early CT Score (ASPECTS)], and collateral circulation deficit volume on CTA.ConclusionMachine learning that is applied to quantifiable image features from CT and CTA alongside basic clinical characteristics constitutes a promising automated method in the pre-interventional prediction of stroke prognosis. Interpretable models allow for exploring which initial features contribute the most to post-thrombectomy outcome prediction overall and for each individual patient outcome.

show abstract

Section: Methodsmentioning

confidence: 99%

Interpretable Machine Learning Modeling for Ischemic Stroke Outcome Prediction

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…In fact, specific diagnostic support software is already used in ischemic stroke in a wide range of hospitals. In this way, this paper presents EffClass, a deep learning model that combines two of the most successful models for image classification and we show that our model is competitive with some of the current most widely used commercial, CE certified and FDA approved methods such as Aidoc [29] or e-ASPECTS [34].…”

Section: Discussionmentioning

confidence: 94%

Deep Learning Applied to Intracranial Hemorrhage Detection

Cortes-Ferre¹,

Gutiérrez–Naranjo²,

Egea-Guerrero³

et al. 2022

Preprint

View full text Add to dashboard Cite

Intracranial hemorrhage is a serious health problem requiring rapid and often intensive medical care. Identifying the location and type of any hemorrhage present is a critical step in treating the patient. Diagnosis requires an urgent procedure and the detection of the hemorrhage is a hard and time-consuming process for human experts. In this paper, we propose a novel method based on Deep Learning techniques which can be useful as decision support system. Our proposal is two-folded. On the one hand, the proposed technique classifies slices of computed tomography scans for hemorrhage existence or not, achieving 92.7% accuracy and 0.978 ROC-AUC. On the other hand, our method provides visual explanation to the chosen classification by using the so-called Grad-CAM method.

show abstract

“…Intracranial LVO identified in the CTA was recorded. Automated and validated ( 14 , 16 ) software was used to generate the ASPECTS (Alberta Stroke Program Early CT Score) and Collateral Score (CS) (Brainomix Ltd., Oxford, UK). Processed images had a maximum slice thickness of 1 mm.…”

Section: Methodsmentioning

confidence: 99%

Automated scoring of collaterals, blood pressure, and clinical outcome after endovascular treatment in patients with acute ischemic stroke and large-vessel occlusion

Guisado‐Alonso

Camps‐Renom²,

Delgado-Mederos³

et al. 2022

Front. Neurol.

View full text Add to dashboard Cite

IntroductionWe aimed to determine whether the degree of collateral circulation is associated with blood pressure at admission in acute ischemic stroke patients treated with endovascular treatment and to determine its prognostic value.MethodsWe evaluated patients with anterior large vessel occlusion treated with endovascular treatment in a single-center prospective registry. We collected clinical and radiological data. Automated and validated software (Brainomix Ltd., Oxford, UK) was used to generate the collateral score (CS) from the baseline single-phase CT angiography: 0, filling of ≤10% of the occluded MCA territory; 1, 11–50%; 2, 51–90%; 3, >90%. When dichotomized, we considered that CS was good (CS = 2–3), or poor (CS = 0–1). We performed bivariate and multivariable ordinal logistic regression analysis to predict CS categories in our population. The secondary outcome was to determine the influence of automated CS on functional outcome at 3 months. We defined favorable functional outcomes as mRS 0–2 at 3 months.ResultsWe included 101 patients with a mean age of 72.1 ± 13.1 years and 57 (56.4%) of them were women. We classified patients into 4 groups according to the CS: 7 patients (6.9%) as CS = 0, 15 (14.9%) as CS = 1, 43 (42.6%) as CS = 2 and 36 (35.6%) as CS = 3. Admission systolic blood pressure [aOR per 10 mmHg increase 0.79 (95% CI 0.68–0.92)] and higher baseline NIHSS [aOR 0.90 (95% CI, 0.84–0.96)] were associated with a worse CS. The OR of improving 1 point on the 3-month mRS was 1.63 (95% CI, 1.10–2.44) favoring a better CS (p = 0.016).ConclusionIn acute ischemic stroke patients with anterior large vessel occlusion treated with endovascular treatment, admission systolic blood pressure was inversely associated with the automated scoring of CS on baseline CT angiography. Moreover, a good CS was associated with a favorable outcome.

show abstract

e-ASPECTS software improves interobserver agreement and accuracy of interpretation of aspects score

Cited by 22 publications

References 19 publications

Interpretable Machine Learning Modeling for Ischemic Stroke Outcome Prediction

Interpretable Machine Learning Modeling for Ischemic Stroke Outcome Prediction

Deep Learning Applied to Intracranial Hemorrhage Detection

Automated scoring of collaterals, blood pressure, and clinical outcome after endovascular treatment in patients with acute ischemic stroke and large-vessel occlusion

Contact Info

Product

Resources

About