High posterior cerebral artery flow predicts ischemia recurrence in patients with internal carotid artery occlusion.

Recurrent stroke is a dreaded complication of symptomatic internal carotid artery occlusion (ICAO). Transcranial Duplex (TCD)-derived increased flow velocity in the ipsilateral posterior cerebral artery (PCA)-P2 segment indicates activated leptomeningeal collateral recruitment and hemodynamic impairment. Leptomeningeal collaterals are pial vascular connections between the anterior and posterior vascular territories. These secondary collateral routes are activated when primary collaterals via the Circle of Willis are insufficient. Our goal was to test the TCD parameter PCA-P2 flow for prediction of ipsilateral ischemia recurrence. We retrospectively analyzed clinical and ultrasound parameters in patients with ICAO. Together with clinical variables, we tested systolic PCA-P2 flow velocity as predictor of a recurrent ischemic event using logistic regression models. Of 111 patients, 13 showed a recurrent ischemic event within the same vascular territory. Increased flow in the ipsilateral PCA-P2 on transcranial ultrasound (median and interquartile range [IQR]: 60 cm/s [IQR 26] vs. 86 cm/s [IQR 41], p = <0.001), as well as previous transient ischemic attack (TIA) and low NIHSS were associated with ischemia recurrence. Combined into one model, accuracy of these parameters to predict recurrent ischemia was 89.2%. Our data suggest that in patients with symptomatic ICAO, flow increases in the ipsilateral PCA-P2 suggest intensified compensatory efforts when other collaterals are insufficient. Together with the clinical variables, this non-invasive and easily assessable duplex parameter detects ICAO patients at particular risk of recurrent ischemia.

Resolving the Smoking Paradox: No Evidence for Smoking-Induced Preconditioning in Large Vessel Occlusion Stroke.

INTRODUCTION: Smoking is an established risk factor for stroke. However, several studies have reported a better outcome after stroke for patients who smoke. According to this "smoking paradox" hypothesis, smoking might promote less severe strokes, higher collateral scores, and smaller infarct cores. METHODS: In this retrospective study, we screened data of 2,980 acute ischemic stroke patients with MCA-M1 occlusion treated with mechanical thrombectomy. Patients were categorized according to smoking status (current, former, or never). We assessed univariate associations between clinical characteristics and smoking status. Subsequently, we used adjusted regression analysis to evaluate associations of smoking with stroke severity on admission (National Institutes of Health Stroke Scale [NIHSS]; primary endpoint), infarct core volume, and collateral status (secondary endpoints). RESULTS: Out of 320 patients, 19.7% (n = 63) were current smokers and 18.8% (n = 60) were former smokers. Admission NIHSS, reperfusion success, and modified Rankin Scale (mRS) after 3-6 months were similar in all groups. Current smokers were younger, more often male and less likely to have atrial fibrillation compared to former and never smokers. In regression analyses, smoking status was neither associated with admission NIHSS (estimate 0.54, 95% confidence interval [CI]: -1.27-2.35, p = 0.557) nor with collateral status (estimate 0.79, 95% CI: 0.44-1.44, p = 0.447) or infarct core volume (estimate -0.69, 95% CI: -15.15-13.77, p = 0.925 for current vs. never smokers). CONCLUSION: We could not confirm the smoking paradox. Our results support the fact that smoking causes stroke at a younger age, highlighting the role of smoking as a modifiable vascular risk factor.

Deep Learning Versus Neurologists: Functional Outcome Prediction in LVO Stroke Patients Undergoing Mechanical Thrombectomy.

BACKGROUND: Despite evolving treatments, functional recovery in patients with large vessel occlusion stroke remains variable and outcome prediction challenging. Can we improve estimation of functional outcome with interpretable deep learning models using clinical and magnetic resonance imaging data? METHODS: In this observational study, we collected data of 222 patients with middle cerebral artery M1 segment occlusion who received mechanical thrombectomy. In a 5-fold cross validation, we evaluated interpretable deep learning models for predicting functional outcome in terms of modified Rankin scale at 3 months using clinical variables, diffusion weighted imaging and perfusion weighted imaging, and a combination thereof. Based on 50 test patients, we compared model performances to those of 5 experienced stroke neurologists. Prediction performance for ordinal (modified Rankin scale score, 0-6) and binary (modified Rankin scale score, 0-2 versus 3-6) functional outcome was assessed using discrimination and calibration measures like area under the receiver operating characteristic curve and accuracy (percentage of correctly classified patients). RESULTS: In the cross validation, the model based on clinical variables and diffusion weighted imaging achieved the highest binary prediction performance (area under the receiver operating characteristic curve, 0.766 [0.727-0.803]). Performance of models using clinical variables or diffusion weighted imaging only was lower. Adding perfusion weighted imaging did not improve outcome prediction. On the test set of 50 patients, binary prediction performance between model (accuracy, 60% [55.4%-64.4%]) and neurologists (accuracy, 60% [55.8%-64.21%]) was similar when using clinical data. However, models significantly outperformed neurologists when imaging data were provided, alone or in combination with clinical variables (accuracy, 72% [67.8%-76%] versus 64% [59.8%-68.4%] with clinical and imaging data). Prediction performance of neurologists with comparable experience varied strongly. CONCLUSIONS: We hypothesize that early prediction of functional outcome in large vessel occlusion stroke patients may be significantly improved if neurologists are supported by interpretable deep learning models.

Deep transformation models for functional outcome prediction after acute ischemic stroke.

In many medical applications, interpretable models with high prediction performance are sought. Often, those models are required to handle semistructured data like tabular and image data. We show how to apply deep transformation models (DTMs) for distributional regression that fulfill these requirements. DTMs allow the data analyst to specify (deep) neural networks for different input modalities making them applicable to various research questions. Like statistical models, DTMs can provide interpretable effect estimates while achieving the state-of-the-art prediction performance of deep neural networks. In addition, the construction of ensembles of DTMs that retain model structure and interpretability allows quantifying epistemic and aleatoric uncertainty. In this study, we compare several DTMs, including baseline-adjusted models, trained on a semistructured data set of 407 stroke patients with the aim to predict ordinal functional outcome three months after stroke. We follow statistical principles of model-building to achieve an adequate trade-off between interpretability and flexibility while assessing the relative importance of the involved data modalities. We evaluate the models for an ordinal and dichotomized version of the outcome as used in clinical practice. We show that both tabular clinical and brain imaging data are useful for functional outcome prediction, whereas models based on tabular data only outperform those based on imaging data only. There is no substantial evidence for improved prediction when combining both data modalities. Overall, we highlight that DTMs provide a powerful, interpretable approach to analyzing semistructured data and that they have the potential to support clinical decision-making.

Citizen Science in Deliberative Systems: Participation, Epistemic Injustice, and Civic Empowerment.

In this paper, we bring together the literature on citizen science and on deliberative democracy and epistemic injustice. We argue that citizen science can be seen as one element of "deliberative systems," as described by Mansbridge et al. But in order to fulfil its democratic potential, citizen science needs to be attentive to various forms of exclusion and epistemic injustice, as analyzed by Fricker, Medina and others. Moreover, to tap the potentials of citizen science from the perspective of deliberative democracy, it needs to move towards a more empowered approach, in which citizens do not only deliver data points, but also, in invited or uninvited settings, participate in discussions about the goals and implications of research. Integrating citizen science into the deliberative systems approach embeds it in a broader framework of democratic theory and suggests the transmission of certain practical strategies (e.g., randomized sampling). It can also contribute to realism about both the potentials and the limits of citizen science. As part of a deliberative system, citizen science cannot, and need not, be the only place in which reforms are necessary for creating stronger ties between science and society and for aligning science with democratic values.

Machine-learning-based outcome prediction in stroke patients with middle cerebral artery-M1 occlusions and early thrombectomy.

BACKGROUND AND PURPOSE: Clinical outcomes vary substantially among individuals with large vessel occlusion (LVO) stroke. A small infarct core and large imaging mismatch were found to be associated with good recovery. The aim of this study was to investigate whether those imaging variables would improve individual prediction of functional outcome after early (<6 h) endovascular treatment (EVT) in LVO stroke. METHODS: We included 222 patients with acute ischemic stroke due to middle cerebral artery (MCA)-M1 occlusion who received EVT. As predictors, we used clinical variables and region of interest (ROI)-based magnetic resonance imaging features. We developed different machine-learning models and quantified their prediction performance according to the area under the receiver-operating characteristic curves and the Brier score. RESULTS: The rate of successful recanalization was 78%, with 54% patients having a favorable outcome (modified Rankin scale score 0-2). Small infarct core was associated with favorable functional outcome. Outcome prediction improved only slightly when imaging was added to patient variables. Age was the driving factor, with a sharp decrease in likelihood of favorable functional outcome above the age of 78 years. CONCLUSIONS: In patients with MCA-M1 occlusion strokes referred to EVT within 6 h of symptom onset, infarct core volume was associated with outcome. However, ROI-based imaging variables led to no significant improvement in outcome prediction at an individual patient level when added to a set of clinical predictors. Our study is in concordance with current practice, where imaging mismatch or collateral readouts are not recommended as factors for excluding patients with MCA-M1 occlusion for early EVT.

Integrating uncertainty in deep neural networks for MRI based stroke analysis.

At present, the majority of the proposed Deep Learning (DL) methods provide point predictions without quantifying the model's uncertainty. However, a quantification of the reliability of automated image analysis is essential, in particular in medicine when physicians rely on the results for making critical treatment decisions. In this work, we provide an entire framework to diagnose ischemic stroke patients incorporating Bayesian uncertainty into the analysis procedure. We present a Bayesian Convolutional Neural Network (CNN) yielding a probability for a stroke lesion on 2D Magnetic Resonance (MR) images with corresponding uncertainty information about the reliability of the prediction. For patient-level diagnoses, different aggregation methods are proposed and evaluated, which combine the individual image-level predictions. Those methods take advantage of the uncertainty in the image predictions and report model uncertainty at the patient-level. In a cohort of 511 patients, our Bayesian CNN achieved an accuracy of 95.33% at the image-level representing a significant improvement of 2% over a non-Bayesian counterpart. The best patient aggregation method yielded 95.89% of accuracy. Integrating uncertainty information about image predictions in aggregation models resulted in higher uncertainty measures to false patient classifications, which enabled to filter critical patient diagnoses that are supposed to be closer examined by a medical doctor. We therefore recommend using Bayesian approaches not only for improved image-level prediction and uncertainty estimation but also for the detection of uncertain aggregations at the patient-level.

Topical Corticosteroid Concerns in Dermatological Outpatients: A Cross-Sectional and Interventional Study.

BACKGROUND/AIMS: Topical corticosteroid concerns (TCC) are an important issue in patients with atopic dermatitis, leading to non-adherence with poor disease control and increased health care costs. However, neither the prevalence of TCC in a more comprehensible dermatological population nor the impact of patient information on topical corticosteroids given by clinicians is known. Therefore, we assessed the prevalence, characteristics, and sources of TCC in a dermatological population and the impact of written and oral patient information on TCC. METHODS: A total of 643 outpatients with various skin diseases answered a 12-item questionnaire while waiting for the doctor's visit. Patients with TCC quantified their concerns on a discrete visual analogue scale before and after patient information, which consisted of written and oral information about topical corticosteroids (TCS) given by dermatologists. RESULTS: The prevalence of TCC was 41.5%, and that of TCC-related non-adherence was 28.3%. TCC was positively associated with age <60 years, female gender, use of complementary and alternative medicine (CAM) and non-physician health care profession. The leading concerns were skin atrophy, systemic effects, and impairment of the immune system. The most frequent sources of TCC were negative reports by media, family, or friends. Both written and oral patient information significantly reduced TCC. The number needed to benefit from patient information was approximately 2. Non-responders were more often female, TCS-inexperienced, and users of CAM with an intermediate level of education. CONCLUSIONS: TCC are highly prevalent in dermatological patients. Patient information may lower TCC in almost every second patient.