Diffusion-weighted imaging (DWI) ischemic volume is related to FLAIR hyperintensity-DWI mismatch and functional outcome after endovascular therapy

Individualized interpretation for the clinical significance of fluid-attenuated inversion recovery vessel hyperintensity in ischemic stroke and transient ischemic attack: A systematic narrative review

Fluid-attenuated inversion recovery (FLAIR) vessel hyperintensity（FVH）refers to the hyperintensity corresponding to the arteries in the subarachnoid space. It is caused by critically slowed blood flow and is commonly encountered in patients with large artery steno-occlusions. Quite a few studies have focused on the clinical significance of FLAIR vessel hyperintensity in terms of its relationship to the prognosis of transient ischemic attack (TIA), baseline severity or infarction volume, early neurological deterioration or infarction growth, and functional outcomes in acute ischemic stroke (AIS). However, inconsistent or conflicting findings were common in these studies and caused confusion in the clinical decision-making process guided by this imaging marker. Through reviewing the available studies on the etiologic mechanism of FVH and investigating findings on its clinical significance in AIS and TIA, this review aims to elucidate the key factors for interpreting the clinical significance of FVH individually.

Development of machine learning models for predicting outcome in patients with distal medium vessel occlusions: a retrospective study

Background

While numerous prognostic factors have been reported for large vessel occlusion (LVO)-acute ischemic stroke (AIS) patients, the same cannot be said for distal medium vessel occlusions (DMVOs). We used machine learning (ML) algorithms to develop a model predicting the short-term outcome of AIS patients with DMVOs using demographic, clinical, and laboratory variables and baseline computed tomography (CT) perfusion (CTP) postprocessing quantitative parameters.

Methods

In this retrospective cohort study, consecutive patients with AIS admitted to two comprehensive stroke centers between January 1, 2017, and September 1, 2022, were screened. Demographic, clinical, and radiological data were extracted from electronic medical records. The clinical outcome was divided into two categories, with a cut-off defined by the median National Institutes of Health Stroke Scale (NIHSS) shift score. Data preprocessing involved addressing missing values through imputation, scaling with a robust scaler, normalization using min-max normalization, and encoding of categorical variables. The data were split into training and test sets (70:30), and recursive feature elimination (RFE) was employed for feature selection. For ML analyses, XGBoost, LightGBM, CatBoost, multi-layer perceptron, random forest, and logistic regression algorithms were utilized. Performance evaluation involved the receiver operating characteristic (ROC) curve, precision-recall curve (PRC), the area under these curves, accuracy, precision, recall, and Matthews correlation coefficient (MCC). The relative weights of predictor variables were examined using Shapley additive explanations (SHAP).

Results

Sixty-nine patients were included and divided into two groups: 35 patients with favorable outcomes and 34 patients with unfavorable outcomes. Utilizing ten selected features, the XGBoost algorithm achieved the best performance in predicting unfavorable outcomes, with an area under the ROC curve (AUROC) of 0.894 and an area under the PRC curve (AUPRC) of 0.756. The SHAP analysis ranked the following features in order of importance for the XGBoost model: mismatch volume, time-to-maximum of the tissue residue function (Tmax) >6 s, diffusion-weighted imaging (DWI) volume, neutrophil-to-platelet ratio (NPR), mean corpuscular volume (MCV), Tmax >10 s, hemoglobin, potassium, hypoperfusion index (HI), and Tmax >8 s.

Conclusions

Our ML models, trained on baseline quantitative laboratory and CT parameters, accurately predicted the short-term outcome in patients with DMVOs. These findings may aid clinicians in predicting prognosis and may be helpful for future research.

Clinically Ineffective Reperfusion After Endovascular Therapy in Acute Ischemic Stroke

Endovascular treatment is a highly effective therapy for acute ischemic stroke due to large vessel occlusion. However, in clinical practice, nearly half of the patients do not have favorable outcomes despite successful recanalization of the occluded artery. This unfavorable outcome can be defined as having clinically ineffective reperfusion. The objective of the review is to describe clinically ineffective reperfusion after endovascular therapy and its underlying risk factors and mechanisms, including initial tissue damage, cerebral edema, the no-reflow phenomenon, reperfusion injury, procedural features, and variations in postprocedural management. Further research is needed to more accurately identify patients at a high risk of clinically ineffective reperfusion after endovascular therapy and to improve individualized periprocedural management strategies, to increase the chance of achieving favorable clinical outcomes.

Clinical features and FLAIR radiomics nomogram for predicting functional outcomes after thrombolysis in ischaemic stroke

Objective

We explored whether radiomics features extracted from diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) images can predict the clinical outcome of patients with acute ischaemic stroke. This study was conducted to investigate and validate a radiomics nomogram for predicting acute ischaemic stroke prognosis.

Methods

A total of 257 patients with acute ischaemic stroke from three clinical centres were retrospectively assessed from February 2019 to July 2022. According to the modified Rankin scale (mRS) at 3 months, the patients were divided into a favourable outcome group (mRS of 0-2) and an unfavourable outcome group (mRS of 3-6). The high-throughput features from the regions of interest (ROIs) within the radiologist-drawn contour by AK software were extracted. We used two feature selection methods, minimum redundancy and maximum (mRMR) and the least absolute shrinkage and selection operator algorithm (LASSO), to select the features. Three radiomics models (DWI, FLAIR, and DWI-FLAIR) were established. A radiomics nomogram with patient characteristics and radiomics signature was built using a multivariate logistic regression model. The performance of the nomogram was evaluated in the test and validation sets. Ultimately, decision curve analysis was implemented to assess the clinical value of the nomogram.

Results

The FLAIR, DWI, and DWI-FLAIR radiomics model exhibited good prediction performance, with area under the curve (AUCs) of 0.922 (95% CI: 0.876-0.968), 0.875 (95% CI: 0.815-0.935), and 0.895 (95% CI: 0.840-0.950). The radiomics nomogram with clinical characteristics including the overall cerebral small vessel disease (CSVD) burden score, hemorrhagic transformation (HT) and admission National Institutes of Health Stroke Scale score (NIHSS) score and the FLAIR Radscore presented good discriminatory potential in the training set (AUC = 0.94; 95% CI: 0.90-0.98) and test set (AUC = 0.94; 95% CI: 0.87-1), which was validated in the validation set 1 (AUC = 0.95; 95% CI: 0.88-1) and validation set 2 (AUC = 0.90; 95% CI: 0.768-1). In addition, it demonstrated good calibration, and decision curve analysis confirmed the clinical value of this nomogram.

Conclusion

This non-invasive clinical-FLIAR radiomics nomogram shows good performance in predicting ischaemic stroke prognosis after thrombolysis.

Combining machine learning with radiomics features in predicting outcomes after mechanical thrombectomy in patients with acute ischemic stroke

BACKGROUND AND OBJECTIVE

Some patients with mechanical thrombectomy will have a poor prognosis. This study establishes a model for predicting the prognosis after mechanical thrombectomy in acute stroke based on diffusion-weighted imaging (DWI) omics characteristics.

METHODS

A total of 260 stroke patients receiving mechanical thrombectomy in our hospital were randomly divided into a training set (n = 182) and a test set (n = 78) in a 7:3 ratio. The regions of interest (ROI) of the imaging features of the DWI infarct area were extracted, and the minimum absolute contraction and selection operator regression model were used to screen the best radiomics features. A support vector machine classifier established the prediction model of the prognosis after mechanical thrombectomy of acute stroke based on the selected features. The prediction efficiency of the model was evaluated by the receiver operating characteristic (ROC) curve.

RESULTS

A total of 1936 radiomic features were extracted, and six features highly correlated with prognosis were screened after dimensionality reduction. Based on the DWI model, the ROC analysis showed that the area under the curve (AUC) for correct prediction in the training and test sets was 0.945 and 0.920, respectively.

CONCLUSION

The model based on the characteristics of radiomics and machine learning has high predictive efficiency for the prognosis of acute stroke after mechanical thrombectomy, which can be used to guide personalized clinical treatment.

Predicting futile recanalization, malignant cerebral edema, and cerebral herniation using intelligible ensemble machine learning following mechanical thrombectomy for acute ischemic stroke

Purpose

To establish an ensemble machine learning (ML) model for predicting the risk of futile recanalization, malignant cerebral edema (MCE), and cerebral herniation (CH) in patients with acute ischemic stroke (AIS) who underwent mechanical thrombectomy (MT) and recanalization.

Methods

This prospective study included 110 patients with premorbid mRS ≤ 2 who met the inclusion criteria. Futile recanalization was defined as a 90-day modified Rankin Scale score >2. Clinical and imaging data were used to construct five ML models that were fused into a logistic regression algorithm using the stacking method (LR-Stacking). We added the Shapley Additive Explanation method to display crucial factors and explain the decision process of models for each patient. Prediction performances were compared using area under the receiver operating characteristic curve (AUC), F1-score, and decision curve analysis (DCA).

Results

A total of 61 patients (55.5%) experienced futile recanalization, and 34 (30.9%) and 22 (20.0%) patients developed MCE and CH, respectively. In test set, the AUCs for the LR-Stacking model were 0.949, 0.885, and 0.904 for the three outcomes mentioned above. The F1-scores were 0.882, 0.895, and 0.909, respectively. The DCA showed that the LR-Stacking model provided more net benefits for predicting MCE and CH. The most important factors were the hypodensity volume and proportion in the corresponding vascular supply area.

Conclusion

Using the ensemble ML model to analyze the clinical and imaging data of AIS patients with successful recanalization at admission and within 24 h after MT allowed for accurately predicting the risks of futile recanalization, MCE, and CH.

Fluid-Attenuated Inversion Recovery Vascular Hyperintensity in Cerebrovascular Disease: A Review for Radiologists and Clinicians

Neuroradiological methods play important roles in neurology, especially in cerebrovascular diseases. Fluid-attenuated inversion recovery (FLAIR) vascular hyperintensity (FVH) is frequently encountered in patients with acute ischemic stroke and significant intracranial arterial stenosis or occlusion. The mechanisms underlying this phenomenon and the clinical implications of FVH have been a matter of debate. FVH is associated with large-vessel occlusion or severe stenosis, as well as impaired hemodynamics. Possible explanations suggested for its appearance include stationary blood and slow antegrade or retrograde filling of the leptomeningeal collateral circulation. However, the prognostic value of the presence of FVH has been controversial. FVH can also be observed in patients with transient ischemic attack (TIA), which may have different pathomechanisms. Its presence can help clinicians to identify patients who have a higher risk of stroke after TIA. In this review article, we aim to describe the mechanism and influencing factors of FVH, as well as its clinical significance in patients with cerebrovascular disease.

Ability of weakly supervised learning to detect acute ischemic stroke and hemorrhagic infarction lesions with diffusion-weighted imaging

BACKGROUND

Gradient-recalled echo (GRE) sequence is time-consuming and not routinely performed. Herein, we aimed to investigate the ability of weakly supervised learning to identify acute ischemic stroke (AIS) and concurrent hemorrhagic infarction based on diffusion-weighted imaging (DWI).

METHODS

First, we proposed spatially locating small stroke lesions in different positions and hemorrhagic infarction lesions by residual neural and visual geometry group networks using weakly supervised learning. Next, we compared the sensitivity and specificity for identifying automatically concurrent hemorrhagic infarction in stroke patients with the sensitivity and specificity of human readings of diffusion and b₀ images to evaluate the performance of the weakly supervised methods. Also, the labeling time of the weakly supervised approach was compared with that of the fully supervised approach.

RESULTS

Data from a total of 1,027 patients were analyzed. The residual neural network displayed a higher sensitivity than did the visual geometry group network in spatially locating the small stroke and hemorrhagic infarction lesions. The residual neural network had significantly greater patient-level sensitivity than did the human readers (98.4% versus 86.2%, P=0.008) in identifying concurrent hemorrhagic infarction with GRE as the reference standard; however, their specificities were comparable (95.4% versus 96.9%, P>0.99). Weak labeling of lesions required significantly less time than did full labeling of lesions (2.667 versus 10.115 minutes, P<0.001).

CONCLUSIONS

Weakly supervised learning was able to spatially locate small stroke lesions in different positions and showed more sensitivity than did human reading in identifying concurrent hemorrhagic infarction based on DWI. The proposed approach can reduce the labeling workload.

Hemodynamics derived from computational fluid dynamics based on magnetic resonance angiography is associated with functional outcomes in atherosclerotic middle cerebral artery stenosis

BACKGROUND

To investigate the relationship between fluid-attenuated inversion recovery (FLAIR) vascular hyperintensity (FVH), hemodynamics, and functional outcome in atherosclerotic middle cerebral artery (MCA) stenosis using a computational fluid dynamics (CFD) model based on magnetic resonance angiography (MRA), according to a modified Rankin Scale (mRS) at 3 months.

METHODS

A total of 120 patients with 50-99% atherosclerotic MCA stenosis were included. The training and internal validation groups were composed of 99 participants and 21 participants, respectively. Demographic, imaging data, and functional outcome (mRS at 3 months) were collected. Hemodynamic parameters were obtained from the CFD model. The FVH score was based on the number of territories where FVH is positive, according to the spatial distribution in the Alberta Stroke Program Early Computed Tomography Score (ASPECTS). The prediction models were constructed according to clinical and hemodynamic parameters using multivariate logistic analysis. The DeLong test compared areas under the curves (AUCs) of the models.

RESULTS

The multivariable logistic regression analysis showed that the National Institute of Health Stroke Scale (NIHSS) at admission, hypertension, hyperlipidemia, the ratio of wall shear stress before treatment (WSSR_before), and difference in the ratio of wall shear stress (WSSR) were independently associated with functional outcome (all P<0.05). In the training group before treatment, the AUC of model 1a (only clinical variables) and 2a (clinical variables with addition of WSSR_before) were 0.750 and 0.802. After treatment, the AUC of model 1b (only clinical variables) and 2b (clinical variables with addition of difference in WSSR) were 0.815 and 0.883, respectively. The AUC of models with hemodynamic parameters was significantly higher than the models based on clinical variables only (all P<0.05, DeLong test). In the internal validation group before treatment, the AUC of the model (clinical variables) was 0.782, and that of the model (clinical variables and WSSR_before) was 0.800. After treatment, the AUC of the model (clinical variables) was 0.833, and that of the model (clinical variables and difference in WSSR) was 0.861. There were no significant differences between the good and the poor functional outcome group concerning FVH_before scores before treatment (0.30±0.81 vs. 0.26±0.97; P=0.321) and FVH_after scores after treatment (0.08±0.39 vs. 0.00±0.00; P=0.244).

CONCLUSIONS

Hemodynamics was associated with functional outcomes in patients with ischemic stroke attributed to atherosclerotic MCA stenosis, while FVH was not. Hemodynamic parameters were of great importance in the prediction models.

The impact of FLAIR vascular hyperintensity on clinical severity and outcome : A retrospective study in stroke patients with proximal middle cerebral artery stenosis or occlusion

BACKGROUND

The clinical significance of fluid-attenuated inversion recovery vascular hyperintensity (FVH) has not been clarified. The aim of this study was to clarify the effects of FVH on the clinical severity and long-term prognosis of patients with proximal middle cerebral artery (MCA) occlusion or severe stenosis.

METHOD

Because their clinical and imaging data is not accessible, we excluded the patients being treated with IV thrombolysis or mechanical thrombectomy. Clinical and imaging characteristics were documented in 282 consecutive AIS patients with proximal MCA occlusion or severe stenosis. We assessed clinical severity using the National Institutes of Health Stroke Scale (NIHSS) score and clinical outcomes using mRS scores. The average time interval between symptom onset and imaging was 16-18 h. The FVH score according to FVH-ASPECTS ranged from 0 to 7, based on the numbers of territories where FVH is positive.

RESULTS

FVH was observed in 235 (83.33%) of the AIS patients. The FVH(+) group tended to have more alcoholics (65 [27.66%] vs 6 [12.77%], P = 0.032), a higher NIHSS score on the 7th day (3 [1-6] vs 2 [1-3], P = 0.039), more instances of early neurological deterioration (END) (27 [11.4%] vs 1 [2.12%], P = 0.05), and more patients with MCA occlusion (94 [40.00%] vs 3 [6.38%]). Among the patients with positive FVH, a high FVH score represented severe clinical impairment (higher NIHSS score on admission [P = 0.009] and 7th day since admission [P = 0.02]) and poor clinical outcomes. Spearman's rank correlations showed that FVH scores were positively correlated with NIHSS scores on admission and NIHSS scores on the 7th day (P = 0.039; P = 0.017, respectively).

CONCLUSION

In patients with proximal middle cerebral artery (MCA) occlusion or stenosis ≥ 70%, a high FVH score represented severe clinical impairment and poor clinical outcomes. In acute ischemic stroke (AIS) patients with proximal MCA occlusion, a high FVH score represented favorable clinical outcomes.