Are the current MRI criteria using the DWI-FLAIR mismatch concept for selection of patients with wake-up stroke to thrombolysis excluding too many patients?

Thrombolysis in wake-up stroke based on MRI mismatch

OBJECTIVES

Wake-up stroke (WUPS) patients can be selected to intravenous thrombolysis (IVT) treatment based on the Magnetic Resonance Imaging (MRI) mismatch concept. However, recent studies suggest the introduction of modified MRI mismatch criteria, allowing IVT in WUPS patients with a partial mismatch.

MATERIAL AND METHODS

WUPS patients treated with IVT in the NOR-TEST trial and consecutively thereafter at Stavanger University Hospital were included in this study. Patient selection for treatment was performed based on the clinical presentation and the MRI DWI/FLAIR mismatch criteria. MRI examinations were reassessed according to the modified DWI-FLAIR mismatch criteria, allowing partial mismatch. Improvement in NIHSS and mRS at 3 months were used to analyze clinical outcome, and the rate of intracranial hemorrhage (ICH) to analyze safety.

RESULTS

78 WUPS patients were treated with IVT. Only 68 of these patients were independent pre-stroke and included in the clinical analysis. When reassessing the MRI examinations, 41 (60 %) were rated as DWI/ FLAIR mismatch, 14 (21 %) as partial mismatch and 13 (19 %) as match. The results show that the patient groups had a mRS score 0-1 at 3 months measured as primary outcome to respectively 27 (65.9 %), 11 (78.6 %) and 8 (61.5 %); (P = 0.629). The mismatch group showed the best clinical improvement (3-points NIHSS reduction, p = 0.005). No ICH was seen in any of the groups.

CONCLUSION

Our study extended the mismatch concept in clinical praxis to treat WUPS patients with partial mismatch, showing the best clinical outcome in the mismatch group.

Multi-grained contrastive representation learning for label-efficient lesion segmentation and onset time classification of acute ischemic stroke

Ischemic lesion segmentation and the time since stroke (TSS) onset classification from paired multi-modal MRI imaging of unwitnessed acute ischemic stroke (AIS) patients is crucial, which supports tissue plasminogen activator (tPA) thrombolysis decision-making. Deep learning methods demonstrate superiority in TSS classification. However, they often overfit task-irrelevant features due to insufficient paired labeled data, resulting in poor generalization. We observed that unpaired data are readily available and inherently carry task-relevant cues, but are less often considered and explored. Based on this, in this paper, we propose to fully excavate the potential of unpaired unlabeled data and use them to facilitate the downstream AIS analysis task. We first analyze the utility of features at the varied grain and propose a multi-grained contrastive learning (MGCL) framework to learn task-related prior representations from both coarse-grained and fine-grained levels. The former can learn global prior representations to enhance the location ability for the ischemic lesions and perceive the healthy surroundings, while the latter can learn local prior representations to enhance the perception ability for semantic relation between the ischemic lesion and other health regions. To better transfer and utilize the learned task-related representation, we designed a novel multi-task framework to simultaneously achieve ischemic lesion segmentation and TSS classification with limited labeled data. In addition, a multi-modal region-related feature fusion module is proposed to enable the feature correlation and synergy between multi-modal deep image features for more accurate TSS decision-making. Extensive experiments on the large-scale multi-center MRI dataset demonstrate the superiority of the proposed framework. Therefore, it is promising that it helps better stroke evaluation and treatment decision-making.

Posterior circulation ischemic stroke: radiomics-based machine learning approach to identify onset time from magnetic resonance imaging

PURPOSE

Posterior circulation ischemic stroke (PCIS) possesses unique features. However, previous studies have primarily or exclusively relied on anterior circulation stroke cases to build machine learning (ML) models for predicting onset time. To date, there is no research reporting the effectiveness and stability of ML in identifying PCIS onset time. We aimed to build diffusion-weighted imaging-based ML models to identify the onset time of PCIS patients.

METHODS

Consecutive PCIS patients within 24 h of definite symptom onset were included (112 in the training set and 49 in the independent test set). Images were processed as follows: volume of interest segmentation, image feature extraction, and feature selection. Five ML models, naïve Bayes, logistic regression, tree ensemble, k-nearest neighbor, and random forest, were built based on the training set to estimate the stroke onset time (binary classification: ≤ 4.5 h or > 4.5 h). Relative standard deviations (RSD), receiver operating characteristic (ROC) curves, and the calibration plot was performed to evaluate the stability and performance of the five models.

RESULTS

The random forest model had the best performance in the test set, with the highest area under the curve (AUC, 0.840; 95% CI: 0.706, 0.974). This model also achieved the highest accuracy, sensitivity, specificity, positive predictive value, and negative predictive value (83.7%, 64.3%, 91.4%, 75.0%, and 86.5%, respectively). Furthermore, the model had high stability (RSD = 0.0094).

CONCLUSION

The PCIS case-based ML model was effective for estimating the symptom onset time and achieved considerably high specificity and stability.

Automated Assessment of the DWI-FLAIR Mismatch in Patients with Acute Ischemic Stroke: Added Value to Routine Clinical Practice

BACKGROUND AND PURPOSE

The DWI-FLAIR mismatch is used to determine thrombolytic eligibility in patients with acute ischemic stroke when the time since stroke onset is unknown. Commercial software packages have been developed for automated DWI-FLAIR classification. We aimed to use e-Stroke software for automated classification of the DWI-FLAIR mismatch in a cohort of patients with acute ischemic stroke and in a comparative analysis with 2 expert neuroradiologists.

MATERIALS AND METHODS

In this retrospective study, patients with acute ischemic stroke who had MR imaging and known time since stroke onset were included. The DWI-FLAIR mismatch was evaluated by 2 neuroradiologists blinded to the time since stroke onset and automatically by the e-Stroke software. After 4 weeks, the neuroradiologists re-evaluated the MR images, this time equipped with automated predicted e-Stroke results as a computer-assisted tool. Diagnostic performances of e-Stroke software and the neuroradiologists were evaluated for prediction of DWI-FLAIR mismatch status.

RESULTS

A total of 157 patients met the inclusion criteria. A total of 82 patients (52%) had a time since stroke onset of ≤4.5 hours. By means of consensus reads, 81 patients (51.5%) had a DWI-FLAIR mismatch. The diagnostic accuracy (area under the curve/sensitivity/specificity) of e-Stroke software for the determination of the DWI-FLAIR mismatch was 0.72/90.0/53.9. The diagnostic accuracy (area under the curve/sensitivity/specificity) for neuroradiologists 1 and 2 was 0.76/69.1/84.2 and 0.82/91.4/73.7, respectively; both significantly (P < .05) improved to 0.83/79.0/86.8 and 0.89/92.6/85.5, respectively, following the use of e-Stroke predictions as a computer-assisted tool. The interrater agreement (κ) for determination of DWI-FLAIR status was improved from 0.49 to 0.57 following the use of the computer-assisted tool.

CONCLUSIONS

This automated quantitative approach for DWI-FLAIR mismatch provides results comparable with those of human experts and can improve the diagnostic accuracies of expert neuroradiologists in the determination of DWI-FLAIR status.

Automatic triaging of acute ischemic stroke patients for reperfusion therapies using Artificial Intelligence methods and multiple MRI features: A review

BACKGROUND

The selection of appropriate treatments for Acute Ischemic Stroke (AIS), including Intravenous (IV) tissue plasminogen activator (tPA) and Mechanical thrombectomy, is a critical aspect of clinical decision-making. Timely treatment is essential, with recommended administration of therapies within 4.5 h of symptom onset. However, patients with unknown Time Since Stroke (TSS), are often excluded from thrombolysis, even if the stroke onset exceeds 6 h. Current clinical guidelines propose using multimodal Magnetic Resonance Imaging (MRI) to assess various mismatches.

METHODS

The review explores the significance of automatic methods based on Artificial Intelligence (AI) algorithms that utilize multiple MRI features to identify patients who are most likely to benefit from acute reperfusion therapies. These AI methods include TSS classification and patient selection for therapies in the late time window (>6 h) using MRI images to provide detailed stroke information.

RESULTS

The review discusses the challenges and limitations in the existing mismatch methods, which may lead to missed opportunities for reperfusion therapy. To address these limitations, AI approaches have been developed to enhance accuracy and support clinical decision-making. These AI methods have shown promising results, outperforming traditional mismatch assessments and providing improved sensitivity and specificity in identifying patients eligible for reperfusion therapies.

DISCUSSION

In summary, the integration of AI algorithms utilizing multiple MRI features has the potential to enhance accuracy, improve patient outcomes, and positively influence the decision-making process in AIS. However, ongoing research and collaboration among clinicians, researchers, and technologists are vital to realize the full potential of AI in optimizing stroke management.

A deep learning analysis of stroke onset time prediction and comparison to DWI-FLAIR mismatch

INTRODUCTION

When time since stroke onset is unknown, DWI-FLAIR mismatch rating is an established technique for patient stratification. A visible DWI lesion without corresponding parenchymal hyperintensity on FLAIR suggests time since onset of under 4.5 h and thus a potential benefit from intravenous thrombolysis. To improve accuracy and availability of the mismatch concept, deep learning might be able to augment human rating and support decision-making in these cases.

METHODS

We used unprocessed DWI and coregistered FLAIR imaging data to train a deep learning model to predict dichotomized time since ischemic stroke onset. We analyzed the performance of Group Convolutional Neural Networks compared to other deep learning methods. Unlabeled imaging data was used for pre-training. Prediction performance of the best deep learning model was compared to the performance of four independent junior and senior raters. Additionally, in cases deemed indeterminable by human raters, model ratings were used to augment human performance. Post-hoc gradient-based explanations were analyzed to gain insights into model predictions.

RESULTS

Our best predictive model performed comparably to human raters. Using model ratings in cases deemed indeterminable by human raters improved rating accuracy and interrater agreement for junior and senior ratings. Post-hoc explainability analyses showed that the model localized stroke lesions to derive predictions.

DISCUSSION

Our analysis shows that deep learning based clinical decision support has the potential to improve the accessibility of the DWI-FLAIR mismatch concept by supporting patient stratification.

Multicentre registration of wake-up stroke in China (MCRWUSC): a protocol for a prospective, multicentre, registry-based cohort study

INTRODUCTION

Wake-up stroke (WUS) is a type of acute ischaemic stroke (AIS) that occurs during sleep with unknown time of symptom onset. The best treatment is usually not suitable for WUS, as thrombolysis is usually provided to patients who had a symptomatic AIS within a definite 4.5 hours, and WUS remains a therapeutic quandary. Efforts to explore the onset time characteristics of patients who had a WUS and the risk factors affecting poor prognosis support a role for providing new insights by performing multicentre cohort study.

METHODS AND ANALYSIS

This multicentre, nationwide prospective registry will include 21 comprehensive stroke centres, with a goal of recruiting 550 patients who had a WUS in China. In this study, clinical data including patient's clinical characteristics, stroke onset time, imaging findings, therapeutic interventions and prognosis (the National Institutes of Health Stroke Scale Score and the modified Rankin Scale Score at different time points) will be used to develop prediction models for stroke onset time and prognostic evaluation using the fast-processing of ischemic stroke software. The purpose of this study is to identify risk factors influencing prognosis, to investigate the relationship between the time when the symptoms are found and the actual onset time and to establish an artificial intelligence-based model to predict the prognosis of patients who had a WUS.

ETHICS AND DISSEMINATION

This study is approved by the ethics committee of Shanghai Pudong Hospital (Shanghai, China) and rest of all participating centres. The findings will be disseminated through peer-reviewed publications and conference presentations.

PROSPERO REGISTRATION NUMBER

ChiCTR2100049133.

Identifying acute ischemic stroke patients within the thrombolytic treatment window using deep learning

BACKGROUND AND PURPOSE

Treatment of acute ischemic stroke is heavily contingent upon time, as there is a strong relationship between time clock and tissue progression. Work has established imaging biomarker assessments as surrogates for time since stroke (TSS), namely, by comparing signal mismatch between diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) imaging. Our goal was to develop an automatic technique for determining TSS from imaging that does not require subspecialist radiology expertise.

METHODS

Using 772 patients (66 ± 9 years, 319 women), we developed and externally evaluated a deep learning network for classifying TSS from MR images and compared algorithm predictions to neuroradiologist assessments of DWI-FLAIR mismatch. Models were trained to classify TSS within 4.5 hours and performance metrics with confidence intervals were reported on both internal and external evaluation sets.

RESULTS

Three board-certified neuroradiologists' DWI-FLAIR mismatch assessments, based on majority vote, yielded a sensitivity of .62, a specificity of .86, and a Fleiss' kappa of .46 when used to classify TSS. The deep learning method performed similarly to radiologists and outperformed previously reported methods, with the best model achieving an average evaluation accuracy, sensitivity, and specificity of .726, .712, and .741, respectively, on an internal cohort and .724, .757, and .679, respectively, on an external cohort.

CONCLUSION

Our model achieved higher generalization performance on external evaluation datasets than the current state-of-the-art for TSS classification. These results demonstrate the potential of automatic assessment of onset time from imaging without the need for expertly trained radiologists.

Imaging in Acute Anterior Circulation Ischemic Stroke: Current and Future

Clinical trials on acute ischemic stroke have demonstrated the clinical effectiveness of revascularization treatments within an appropriate time window after stroke onset: intravenous thrombolysis (NINDS and ECASS-III) through the administration of tissue plasminogen activator within a 4.5-hour time window, endovascular thrombectomy (ESCAPE, REVASCAT, SWIFT-PRIME, MR CLEAN, EXTEND-IA) within a 6-hour time window, and extending the treatment time window up to 24 hours for endovascular thrombectomy (DAWN and DEFUSE 3). However, a substantial number of patients in these trials were ineligible for revascularization treatment, and treatments of some patients were considerably futile or sometimes dangerous in the clinical trials. Guidelines for the early management of patients with acute ischemic stroke have evolved to accept revascularization treatment as standard and include eligibility criteria for the treatment. Imaging has been crucial in selecting eligible patients for revascularization treatment in guidelines and clinical trials. Stroke specialists should know imaging criteria for revascularization treatment. Stroke imaging studies have demonstrated imaging roles in acute ischemic stroke management as follows: 1) exclusion of hemorrhage and stroke mimic disease, 2) assessment of salvageable brain, 3) localization of the site of vascular occlusion and thrombus, 4) estimation of collateral circulation, and 5) prediction of acute ischemic stroke expecting hemorrhagic transformation. Here, we review imaging methods and criteria to select eligible patients for revascularization treatment in acute anterior circulation stroke, focus on 2019 guidelines from the American Heart Association/American Stroke Association, and discuss the future direction of imaging-based patient selection to improve treatment effects.

MRI radiomic features-based machine learning approach to classify ischemic stroke onset time

PURPOSE

We aimed to investigate the ability of MRI radiomics features-based machine learning (ML) models to classify the time since stroke onset (TSS), which could aid in stroke assessment and treatment options.

METHODS

This study involved 84 patients with acute ischemic stroke due to anterior circulation artery occlusion (51 in the training cohort and 33 in the independent test cohort). Region of infarct segmentation was manually outlined by 3D-slicer software. Image processing including registration, normalization and radiomics features calculation were done in R (version 3.6.1). A total of 4312 radiomic features from each image sequence were captured and used in six ML models to estimate stroke onset time for binary classification (≤ 4.5 h). Receiver-operating characteristic curve (ROC) and other parameters were calculated to evaluate the performance of the models in both training and test cohorts.

RESULTS

Twelve radiomics and six clinic features were selected to construct the ML models for TSS classification. The deep learning model-based DWI/ADC radiomic features performed the best for binary TSS classification in the independent test cohort, with an AUC of 0.754, accuracy of 0.788, sensitivity of 0.952, specificity of 0.500, positive predictive value of 0.769, and negative predictive value of 0.857, respectively. Furthermore, adding clinical information did not improve the performance of the DWI/ADC-based deep learning model. The TSS prediction models can be visited at: http://123.57.65.199:3838/deeptss/ .

CONCLUSIONS

A unique deep learning model based on DWI/ADC radiomic features was constructed for TSS classification, which could aid in decision making for thrombolysis in patients with unknown stroke onset.

Time-Based Decision Making for Reperfusion in Acute Ischemic Stroke

Decision making in the extended time windows for acute ischemic stroke can be a complex and time-consuming process. The process of making the clinical decision to treat has been compounded by the availability of different imaging modalities. In the setting of acute ischemic stroke, time is of the essence and chances of a good outcome diminish by each passing minute. Navigating the plethora of advanced imaging modalities means that treatment in some cases can be inefficaciously delayed. Time delays and individually based non-programmed decision making can prove challenging for clinicians. Visual aids can assist such decision making aimed at simplifying the use of advanced imaging. Flow charts are one such visual tool that can expedite treatment in this setting. A systematic review of existing literature around imaging modalities based on site of occlusion and time from onset can be used to aid decision making; a more program-based thought process. The use of an acute reperfusion flow chart helping navigate the myriad of imaging modalities can aid the effective treatment of patients.

Clinical feasibility of single-shot fluid-attenuated inversion recovery with wide inversion recovery pulse designed to reduce cerebrospinal fluid and motion artifacts for evaluation of uncooperative patients in acute stroke protocol

Fluid-attenuated inversion recovery (FLAIR) imaging is a key sequence for stroke assessment. Motion artifact reduction with short acquisition time is still challenging, but necessary in the magnetic resonance (MR) stroke protocol, especially for uncooperative patients suspected of stroke. The aim of this study is to investigate the feasibility of modified single-shot FLAIR with wide inversion recovery pulses for use in stroke patients. This is a prospective study, which included 30 patients clinically suspected of stroke who were examined by MR stroke protocol from January 2018 to September 2018. A 1.5 T, multi-shot-turbo spin-echo (TSE) conventional FLAIR, and single-shot-TSE-FLAIR with wide inversion recovery pulse were used. Modified single-shot FLAIR was obtained for 30 patients with suspected stroke who moved during conventional FLAIR scan. Motion artifacts were randomly and independently scored using a 5-grade scale by three radiologists in blinded fashion. Whether the FLAIR vessel hyperintensity sign was present was visually evaluated. Statistical tests included Wilcoxon-signed rank test and weighted Cohen's kappa statistics. The motion artifact score was significantly lower in single-shot FLAIR than in conventional FLAIR (0.37 ± 0.56 vs. 1.83 ± 1.18; p < 0.05. The vessel hyperintensity sign was visualized in 6 and 5 patients on single-shot and conventional FLAIR images, respectively. This study demonstrates the value of single-shot FLAIR for stroke assessment. Single-shot FLAIR reduced motion artifact and visualized vessel hyperintensity sign more than conventional FLAIR. LEVEL OF EVIDENCE: 2. TECHNICAL EFFICACY STAGE: 2.

Clinical outcomes and safety profile of Tenecteplase in wake-up stroke

BACKGROUND

Tenecteplase has probably pharmacological and clinical advantages in the treatment of acute ischemic stroke. There are lacking data about safety and efficacy of tenecteplase in wake-up stroke (WUPS).

AIMS

To investigate safety and efficacy of tenecteplase compared to alteplase in WUPS patients included in NOR-TEST.

METHODS

WUPS patients in NOR-TEST were included in the study based on DWI-FLAIR mismatch. Included patients randomly assigned (1:1) to receive intravenous tenecteplase 0.4 mg/kg (to a maximum of 40 mg) or alteplase 0.9 mg/kg (to a maximum of 90 mg). Neurological improvement was defined as 1) favorable functional outcome at 90 days modified Rankin Scale (mRS) of 0 or 1 and 2) neurological improvement measured with the National Institutes of Health Stroke Scale (NIHSS) of 4 points within 24 hours as compared to admission NIHSS or NIHSS 0 at 24 hours.

RESULTS

Of 1100 patients from 13 stroke centers included in NOR-TEST, 45 were WUPS patients. Of these, 5 patients were stroke mimics and excluded. Of the remaining 40 patients (3.6%), 24 were treated with alteplase (60%). There was no difference in the number of patients achieving a good clinical outcome (mRS 0-1) in either treatment group. Patients treated with tenecteplase showed a better early neurological improvement (87.5% vs 54.2%, P = 0.027). No ICH was detected on MRI/CT 24-28 hours after thrombolysis.

CONCLUSIONS

In WUPS patients treated in NOR-TEST, there was no difference in clinical outcomes at 90 days and no ICH events or deaths were observed in either alteplase- or tenecteplase-treated patients. Clinical Trial Registration-URL: https://www.clinicaltrials.gov. Unique identifier: NCT01949948.

Pathophysiologic mechanisms, neuroimaging and treatment in wake-up stroke

Wake-up stroke (WUS) or ischemic stroke occurring during sleep accounts for 14%-29.6% of all ischemic strokes. Management of WUS is complicated by its narrow therapeutic time window and attributable risk factors, which can affect the safety and efficacy of administering intravenous (IV) tissue plasminogen activator (t-PA). This manuscript will review risk factors of WUS, with a focus on obstructive sleep apnea, potential mechanisms of WUS, and evaluate studies assessing safety and efficacy of IV t-PA treatment in WUS patients guided by neuroimaging to estimate time of symptom onset. The authors used PubMed (1966 to March 2018) to search for the term "Wake-Up Stroke" cross-referenced with "pathophysiology," ''pathogenesis," "pathology," "magnetic resonance imaging," "obstructive sleep apnea," or "treatment." English language Papers were reviewed. Also reviewed were pertinent papers from the reference list of the above-matched manuscripts. Studies that focused only on acute Strokes with known-onset of symptoms were not reviewed. Literature showed several potential risk factors associated with increased risk of WUS. Although the onset of WUS is unknown, a few studies investigated the potential benefit of magnetic resonance imaging (MRI) in estimating the age of onset which encouraged conducting clinical trials assessing the efficacy of MRI-guided thrombolytic therapy in WUS.

Wake-up stroke: imaging-based diagnosis and recanalization therapy

Wake-up stroke (WUS) is a subgroup of ischemic stroke in which patients show no abnormality before sleep while wake up with neurological deficits. In addition to the uncertain onset, WUS patients have difficulty to receive prompt and effective thrombolytic or reperfusion therapy, leading to relatively poor prognosis. A number of researches have indicated that CT or MRI based thrombolysis and endovascular therapy might have benefits for WUS patients. This review article narratively discusses the pathogenesis, risk factors, imaging-based diagnosis and recanalization treatments of WUS with the purpose of expanding current treatment options for this group of stroke patients and exploring better therapeutic methods. The result showed that multimodal MRI or CT scan might be the best methods for extending the time window of WUS and, therefore, a large proportion of WUS patients could have favorable prognosis.

CT radiomics features as a diagnostic tool for classifying basal ganglia infarction onset time

OBJECTIVE

This study was aimed to discuss the application of radiomics using CT analysis in basal ganglia infarction (BGI) for determining the time since stroke onset (TSS) which could provide critical information to clinicians in deciding stroke treatment options such as thrombolysis.

METHODS

This study involved 316 patients with BGI (237 in the training cohort and 79 in the independent validation cohort). Region of interest segmentation and feature extraction was done by ITK-SNAP software. We used the existing medical history to binarize the TSS into two categories: positive (< 4.5 h) and negative (≥ 4.5 h). The key radiomic signature features were retrieved by the least absolute shrinkage and selection operator multiple logistic regression model. Receiver operating characteristic curve and AUC analysis were used to evaluate the performance of the radiomic signature in both the training and validation cohorts.

RESULTS

295 features were extracted from a manually outlined infarction region. Five features were selected to construct the radiomic signature for TSS classification purposes. The performance of the radiomic signature to distinguish between positive and negative in the training cohort was good, with an AUC of 0.982, a sensitivity of 0.929, and a specificity of 0.959. In the validation cohort, the radiomic signature showed an AUC of 0.974, a sensitivity of 0.951, and a specificity of 0.961.

CONCLUSION

A unique radiomic signature was constructed for use as a diagnostic tool for discriminating the TSS in BGI and may guide decisions to use thrombolysis in patients with unknown times of BGI onset.

Machine Learning Approach to Identify Stroke Within 4.5 Hours

Background and Purpose- We aimed to investigate the ability of machine learning (ML) techniques analyzing diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) magnetic resonance imaging to identify patients within the recommended time window for thrombolysis. Methods- We analyzed DWI and FLAIR images of consecutive patients with acute ischemic stroke within 24 hours of clear symptom onset by applying automatic image processing approaches. These processes included infarct segmentation, DWI, and FLAIR imaging registration and image feature extraction. A total of 89 vector features from each image sequence were captured and used in the ML. Three ML models were developed to estimate stroke onset time for binary classification (≤4.5 hours): logistic regression, support vector machine, and random forest. To evaluate the performance of ML models, the sensitivity and specificity for identifying patients within 4.5 hours were compared with the sensitivity and specificity of human readings of DWI-FLAIR mismatch. Results- Data from a total of 355 patients were analyzed. DWI-FLAIR mismatch from human readings identified patients within 4.5 hours of symptom onset with 48.5% sensitivity and 91.3% specificity. ML algorithms had significantly greater sensitivities than human readers (75.8% for logistic regression, P=0.020; 72.7% for support vector machine, P=0.033; 75.8% for random forest, P=0.013) in detecting patients within 4.5 hours, but their specificities were comparable (82.6% for logistic regression, P=0.157; 82.6% for support vector machine, P=0.157; 82.6% for random forest, P=0.157). Conclusions- ML algorithms using multiple magnetic resonance imaging features were feasible even more sensitive than human readings in identifying patients with stroke within the time window for acute thrombolysis.

A Machine Learning Approach for Classifying Ischemic Stroke Onset Time From Imaging

Current clinical practice relies on clinical history to determine the time since stroke (TSS) onset. Imaging-based determination of acute stroke onset time could provide critical information to clinicians in deciding stroke treatment options, such as thrombolysis. The patients with unknown or unwitnessed TSS are usually excluded from thrombolysis, even if their symptoms began within the therapeutic window. In this paper, we demonstrate a machine learning approach for TSS classification using routinely acquired imaging sequences. We develop imaging features from the magnetic resonance (MR) images and train machine learning models to classify the TSS. We also propose a deep-learning model to extract hidden representations for the MR perfusion-weighted images and demonstrate classification improvement by incorporating these additional deep features. The cross-validation results show that our best classifier achieved an area under the curve of 0.765, with a sensitivity of 0.788 and a negative predictive value of 0.609, outperforming existing methods. We show that the features generated by our deep-learning algorithm correlate with the MR imaging features, and validate the robustness of the model on imaging parameter variations (e.g., year of imaging). This paper advances magnetic resonance imaging analysis one-step-closer to an operational decision support tool for stroke treatment guidance.

Modified DWI-FLAIR mismatch guided thrombolysis in unknown onset stroke

DWI-FLAIR mismatch has been recently proven to identify patients with unknown onset stroke (UOS) eligible for thrombolysis. However, this concept may exclude patients from thrombolysis who may eventually benefit as well. We aimed to examine the feasibility, safety and potential efficacy of thrombolysis in wake-up stroke (WUS) and UOS patients using a modified DWI-FLAIR mismatch allowing for partial FLAIR positivity. WUS/UOS patients fulfilling the modified DWI-FLAIR mismatch and treated with intravenous thrombolysis (IVT) were compared to propensity score matched WUS/UOS patients excluded from IVT due to FLAIR positivity. The primary endpoint was a symptomatic intracranial hemorrhage (SICH), the secondary endpoints were improvement of ≥ 4 in NIHSS score and mRS score at 3 months. 64 IVT-treated patients (median NIHSS 9) and 64 controls (median NIHSS 8) entered the analysis (p = 0.2). No significant difference in SICH was found between the IVT group and the controls (3.1% vs. 1.6%, p = 0.9). An improvement of ≥ 4 NIHSS points was more frequent in IVT patients as compared to controls (40.6% vs. 18.8%, p = 0.01). 23.4% of IVT patients achieved a mRS score of 0-1 at 3 months as compared to 18.8% of the controls (p = 0.8). SICH, improvement of NIHSS ≥ 4 and mRS 0-1 at 3 months were comparable in thrombolyzed patients with negative FLAIR images versus those thrombolyzed with partial positive FLAIR images (3% vs. 3%, p = 0.9; 40% vs. 41%, p = 0.9; 19% vs. 22%, p = 0.8). Our study signalizes that thrombolysis may be feasible in selected WUS/UOS patients with partial FLAIR signal positivity.

Mapping a Reliable Stroke Onset Time Course Using Signal Intensity on DWI Scans

BACKGROUND AND PURPOSE

Identifying a last known well (LKW) time surrogate for acute stroke is vital to increase stroke treatment. Diffusion-weighted imaging (DWI) signal intensity initially increases from onset of stroke but mapping a reliable time course to the signal intensity has not been demonstrated.

METHODS

We retrospectively reviewed stroke code patients between 1/2016 and 6/2017 from the prospective; Institutional review board (IRB) approved University of California San Diego Stroke Registry. Patients who had magnetic resonance imaging of brain from onset, with or without intervention, are included. All ischemic strokes were confirmed and timing from onset to imaging was calculated. Raw DWI intensity is measured using IMPAX software and compared to contralateral side for control for a relative DWI intensity (rDWI). LKW and magnetic resonance imaging (MRI) time were collected by chart review. Correlation is assessed using Pearson correlation coefficient between DWI intensity, rDWI, and time to MRI imaging. 1.5T, 3T, and combined modalities were examined.

RESULTS

Seventy-eight patients were included in this analysis. Overall, there was statistically significant positive correlation (.53, P < .001) between DWI intensity and LKW time irrespective of scanner strength. Using 1.5T analyses, there was good correlation (.46, P < .001). 3T MRI analysis further showed comparatively stronger positive correlation (.66, P < .001).

CONCLUSIONS

There is good correlation between DWI intensity and minutes from onset to MRI. This suggests a time-dependent DWI intensity response and supports the potential use of DWI intensity measurements to extrapolate an LKW time. Further studies are being pursued to increase both experience and generalizability.

Safety and cost-effectiveness thrombolysis by diffusion-weighted imaging and fluid attenuated inversion recovery mismatch for wake-up stroke

Wake-up stroke, defined as patients who wake up with stroke symptoms which were not present prior to falling asleep, accounted for 14%-25% of acute ischemic stroke. Due to the unknown time of symptom onset, wake-up stoke was not in including criteria of intravenous thrombolysis. Several large randomized stroke trials using diffusion-weighted imaging(DWI)and fluid attenuated inversion recovery(FLAIR)mismatch patient selection may identify a subset of patients with wake-up stroke that can safely and effectively benefit from intravenous thrombolysis. In addition, economic factor was another important limitation to generalize thrombolysis treatment. Fortunately, MRI-based thrombolysis was a cost-effective treatment for wake-up stroke compared to these patients with no thrombolysis.

Classifying Acute Ischemic Stroke Onset Time using Deep Imaging Features

Models have been developed to predict stroke outcomes (e.g., mortality) in attempt to provide better guidance for stroke treatment. However, there is little work in developing classification models for the problem of unknown time-since-stroke (TSS), which determines a patient's treatment eligibility based on a clinical defined cutoff time point (i.e., <4.5hrs). In this paper, we construct and compare machine learning methods to classify TSS<4.5hrs using magnetic resonance (MR) imaging features. We also propose a deep learning model to extract hidden representations from the MR perfusion-weighted images and demonstrate classification improvement by incorporating these additional imaging features. Finally, we discuss a strategy to visualize the learned features from the proposed deep learning model. The cross-validation results show that our best classifier achieved an area under the curve of 0.68, which improves significantly over current clinical methods (0.58), demonstrating the potential benefit of using advanced machine learning methods in TSS classification.

Wake-up stroke-Amendable for thrombolysis-like stroke with known onset time?

Patients suffering an acute ischemic stroke can be treated with intravenous thrombolysis in the absence of contraindications. A known onset time is a prerequisite as treatment, according to guidelines, has to be started within 4.5 hours. In patients awakening with a stroke, the last time they were seen without a neurological deficit is assumed to be the time of onset. Thus, despite of lack of contraindications on initial brain imaging, these patients are largely excluded from therapy. This review discusses the underlying pathophysiological, clinical, and radiological evidence surrounding wake-up stroke and its consequences for making treatment decisions.

Diffusion-weighted magnetic resonance imaging reflects activation of signal transducer and activator of transcription 3 during focal cerebral ischemia/reperfusion

Signal transducer and activator of transcription (STAT) is a unique protein family that binds to DNA, coupled with tyrosine phosphorylation signaling pathways, acting as a transcriptional regulator to mediate a variety of biological effects. Cerebral ischemia and reperfusion can activate STATs signaling pathway, but no studies have confirmed whether STAT activation can be verified by diffusion-weighted magnetic resonance imaging (DWI) in rats after cerebral ischemia/reperfusion. Here, we established a rat model of focal cerebral ischemia injury using the modified Longa method. DWI revealed hyperintensity in parts of the left hemisphere before reperfusion and a low apparent diffusion coefficient. STAT3 protein expression showed no significant change after reperfusion, but phosphorylated STAT3 expression began to increase after 30 minutes of reperfusion and peaked at 24 hours. Pearson correlation analysis showed that STAT3 activation was correlated positively with the relative apparent diffusion coefficient and negatively with the DWI abnormal signal area. These results indicate that DWI is a reliable representation of the infarct area and reflects STAT phosphorylation in rat brain following focal cerebral ischemia/reperfusion.

The Effects of DWI-Infarct Lesion Volume on DWI-FLAIR Mismatch: Is There a Need for Size Stratification?

BACKGROUND

The lack of fluid-attenuated inversion-recovery (FLAIR) hyperintensity in areas of diffusion-weighted imaging (DWI) high signal, or DWI-FLAIR mismatch, is a potential imaging biomarker for timing of stroke onset. We aimed to determine the effects of DWI infarct lesion volume on DWI-FLAIR mismatch and its accuracy for identification of strokes within intravenous (IV) the thrombolytic therapy window.

METHODS

Acute ischemic stroke patients with magnetic resonance imaging scan within 12 hours of witnessed stroke were included. Two neuroradiologists independently reviewed DWI and FLAIR sequences for DWI-FLAIR mismatch in areas of restricted diffusion compared to the contralateral normal side.

RESULTS

DWI-FLAIR mismatch was identified in 21/82 (26%) patients. Infarct lesions with DWI-FLAIR mismatch were scanned earlier (3.8 ± .3 vs. 7.5 ± .3 hours from onset, P < .001) and were smaller in size (8.9±2.3 vs. 43.1±11.9 mL, P = .007) compared to lesions without mismatch. Multivariate regression analysis showed a significant interaction between lesion volume and time-from-onset in relationship with the presence of DWI-FLAIR mismatch (P = .045). The presence of DWI-FLAIR mismatch had 56% sensitivity, 83% specificity, 48% positive predictive value (PPV), and 87% negative predictive value (NPV) for identification of infarction within 4.5 hours of symptom onset; while for infarct lesions >15 mL, the DWI-FLAIR mismatch had 100% specificity and PPV for acute infarcts within 4.5 hours of onset.

CONCLUSION

The effects of stroke onset-to-scan time gap on DWI-FLAIR mismatch are not the same for different DWI lesion volumes. At DWI lesion volumes >15 mL, the DWI-FLAIR mismatch is highly specific for acute infarcts within IV thrombolytic therapy time, and can identify wake-up stroke patients eligible for treatment.

Wake-up stroke: Clinical characteristics, sedentary lifestyle, and daytime sleepiness

Summary Objective: Wake-up stroke (WUS) is defined when the exact time of the beginning of the symptoms cannot be determined, for the deficits are perceived upon awakening. Sleep alterations are important risk factors for stroke and cardiovascular diseases. This study evaluates the characteristics of patients with and without WUS, the presence of daytime sleepiness, and associated risk factors. Method: Patients with ischemic stroke were investigated about the presence of WUS. Clinical and demographic characteristics were evaluated. Stroke severity was studied by the National Institutes of Health Stroke Scale (NIHSS) and the Modified Rankin Scale (MRS), and daytime sleepiness severity was studied by the Epworth Sleepiness Scale (ESS). Results: Seventy patients (57.1% men) aged from 32 to 80 years (58.5±13.3) were studied. WUS was observed in 24.3%. Arterial hypertension (67.1%), type 2 diabetes (27.1%), and hyperlipidemia (22.8%) were frequent. Type 2 diabetes and sedentary lifestyle were more common in patients with WUS (p<0.05). Overall, mild, moderate or very few symptoms of stroke (NIHSS<5) were predominant (62.3%). Among all cases, 20% had excessive daytime sleepiness (ESS>10). No differences were found between patients with and without WUS as regards stroke severity or excessive daytime sleepiness. Patients with excessive daytime sleepiness were younger and had more sedentary lifestyle (p<0.05). Individuals with previous history of heavy drinking had more daytime sleepiness (p=0.03). Conclusion: Wake-up stroke occurs in approximately 25% of stroke cases. In this study, patients with WUS had more diabetes and sedentary lifestyle. Daytime sleepiness is frequent and is associated with sedentary lifestyle and heavy drinking.

Radiological imaging in acute ischaemic stroke

Patients who suffer acute ischaemic stroke can be treated with thrombolysis if therapy is initiated early. Radiological evaluation of the intracranial tissue before such therapy can be given is mandatory. In this review current radiological diagnostic strategies are discussed for this patient group. Beyond non-enhanced computed tomography (CT), the standard imaging method for many years, more sophisticated CT stroke protocols including CT angiography and CT perfusion have been developed, and additionally an increasing number of patients are examined with magnetic resonance imaging as the first imaging method used. Advantages and challenges of the different methods are discussed.