Predictor Of Stroke Research Articles

Non-Cardiac and Cardiac Risk for Ischemic Stroke in Young Adults: The Stanford Y-CORE (Young Cardiovascular Outcomes and Risk Evaluation) Study.

Stroke incidence is increasing in young to middle-aged adults. Assessing risk factors is important in this large population segment whose comorbidities may differ from older adults. In this retrospective cohort analysis of adults aged between 20-50 presenting to the Stanford Healthcare system from January 1, 2000, through December 31, 2021, with no prior history of stroke or transient ischemic attack, we studied the effects of 30 risk factors on the primary endpoint of incident ischemic stroke, defined by the presence of ICD-10 codes for stroke and confirmed by brain imaging. The secondary endpoint was incident cerebrovascular events defined by the presence of ICD-10 codes for stroke or transient ischemic attacks (TIA). Associations were measured with time-varying multivariable survival regression. From an overall population of 1.3 million, we identified 540,999 individuals aged 20-50Y. Over the study period, 802 experienced the primary endpoint and 5,734 the secondary endpoint. On multivariable analysis, non-cardiovascular risk factors were independently associated with the primary endpoint adjusting for established cardiovascular risk factors, including sleep apnea [1.44, (1.19, 1.74)], bipolar disorder [1.88, (1.23, 2.86)], cancer [2.07 (1.71, 2.51)], chronic kidney disease (CKD) [2.2 (1.73, 2.81)]. Other non-cardiovascular associations included ethno-racial subgroups of Black [2.05, (1.60, 2.64)], Pacific Islander [2.56, (1.70, 3.84)]; and Hispanic [1.71, (1.37, 2.15)] versus white non-Hispanics. Combining non-cardiovascular risk factors significant on multivariable analysis with established cardiovascular factors significantly improved the C-index for de novo stoke to 0.814 over that obtained in either group alone (P<0.05). In this large population of young adults, several non-cardiovascular factors conferred risk for incident stroke independent of known cardiovascular risk factors and, in combination, significantly improved the prediction of incident stroke over those based on either group of factors alone. These findings may have implications for assessing risk in younger patients with distinct comorbidities.

Predicting cerebral infarction and transient ischemic attack in healthy individuals and those with dysmetabolism: a machine learning approach combined with routine blood tests

Ischemic cerebral infarction is the most prevalent type of stroke, causing significant disability and death worldwide. Transient ischemic attack (TIA) is a strong predictor of subsequent stroke. Individuals with dysmetabolism, such as hypertension, hypercholesterolemia, and diabetes, are at increased risk for cerebral infarction (CI) and TIA. In resource-limited settings, diagnosing CI and TIA can be particularly difficult due to a lack of advanced imaging and specialized expertise. Therefore, we aim to develop a simple, convenient, blood-based approach that could assist clinicians in diagnosing CI and TIA, especially in regions where advanced imaging or stroke-specific expertise is limited. All study subjects were patients admitted to the First Hospital of Xiamen University and healthy check-up populations between January 2018 and September 2023. This study employed five machine learning methods alongside 21 blood routine indicators, 30 blood biochemical indicators, age, and gender to construct predictive models for CI and TIA in both healthy individuals and those with dysmetabolism. The Area Under the Receiver Operating Characteristic (ROC) Curve (AUC) served as the metric to assess the comprehensive predictive capability of the models. Subsequently, the SHAP package was employed for model interpretation. Extreme Gradient Boosting (XGBoost) outperforms other models in all predictive models. In the models predicting CI and TIA among healthy, the AUC is 0.9958 (0.9947–0.9969) and 0.9928 (0.9899–0.9951), respectively. Among the nine shared key features of the two models are indicators of glucose metabolism, lipid metabolism, and liver metabolism. In the models for predicting CI and TIA among patients with hypertension, hypercholesterolemia, diabetes, and those with all three metabolic disorders combined, the AUCs ranged from 0.6990 to 0.8591. We found that the indicators K significantly contributed to predict CI and TIA from those with dysmetabolism. Additionally, metabolic-related indicators, such as glucose (GLU) and high-density lipoprotein cholesterol (HDL-C), are ranked highly among the top ten contributing features. XGBoost performed the best in all models. It can effectively differentiate CI and TIA from healthy and dysmetabolic patients by combining blood routine and blood biochemical indicators. Moreover, it can also differentiate CI from TIA. Although any suspicious findings from this model would still require confirmatory imaging, the simplicity and low cost of blood-based testing may offer a practical adjunct for clinicians—particularly in areas lacking advanced imaging or extensive stroke expertise—and could facilitate earlier diagnostic decision-making.

A different perspective on studying stroke predictors: joint models for longitudinal and time-to-event data in a type 2 diabetes mellitus cohort

BackgroundMost predictive models rely on risk factors and clinical outcomes assessed simultaneously. This approach does not adequately reflect the progression of health conditions. By employing joint models of longitudinal and survival data, we can dynamically adjust prognosis predictions for individual patients. Our objective was to optimize the prediction of stroke or transient ischemic attack (TIA) via joint models that incorporate all available changes in the predictive variables.MethodsA total of 3442 patients with type 2 diabetes mellitus (T2DM) and no history of stroke, TIA or myocardial infarction were followed for 12 years. Models were constructed independently for men and women. We used proportional hazards regression models to assess the effects of baseline characteristics (excluding longitudinal data) on the risk of stroke/TIA and linear mixed effects models to assess the effects of baseline characteristics on longitudinal data development over time. Both submodels were then combined into a joint model. To optimize the analysis, a univariate analysis was first performed for each longitudinal predictor to select the functional form that gave the best fit via the deviance information criterion. The variables were then entered into a multivariate model using pragmatic criteria, and if they improved the discriminatory ability of the model, the area under the curve (AUC) was used.ResultsDuring the follow-up period, 303 patients (8.8%) experienced their first stroke/TIA. Age was identified as an independent predictor among males. Among females, age was positively associated with atrial fibrillation (AF). The final model for males included AF, systolic blood pressure (SBP), and diastolic blood pressure (DBP), with albuminuria and the glomerular filtration rate (GFR) as adjustment variables. For females, the model included AF, blood pressure (BP), and renal function (albuminuria and GFR), with HbA1c and LDL cholesterol as adjustment variables. Both models demonstrated an AUC greater than 0.70.ConclusionsAge, AF, and SBP have been confirmed as significant predictive factors in both sexes, whereas renal function was significant only in women. Interestingly, an increase in DBP may serve as a protective factor in our cohort. These factors were particularly relevant in the last 3–7 years of follow-up.

A novel intelligent healthcare system based on efficient deep learning models for brain stroke prediction

A novel intelligent healthcare system based on efficient deep learning models for brain stroke prediction

Artificial Intelligence in Atrial Fibrillation: From Early Detection to Precision Therapy

Atrial fibrillation (AF) is the most prevalent cardiac arrhythmia, associated with significant morbidity, mortality, and healthcare burden. Despite advances in AF management, challenges persist in early detection, risk stratification, and treatment optimization, necessitating innovative solutions. Artificial intelligence (AI) has emerged as a transformative tool in AF care, leveraging machine learning and deep learning algorithms to enhance diagnostic accuracy, improve risk prediction, and guide therapeutic interventions. AI-powered electrocardiographic screening has demonstrated the ability to detect asymptomatic AF, while wearable photoplethysmography-based technologies have expanded real-time rhythm monitoring beyond clinical settings. AI-driven predictive models integrate electronic health records and multimodal physiological data to refine AF risk stratification, stroke prediction, and anticoagulation decision making. In the realm of treatment, AI is revolutionizing individualized therapy and optimizing anticoagulation management and catheter ablation strategies. Notably, AI-enhanced electroanatomic mapping and real-time procedural guidance hold promise for improving ablation success rates and reducing AF recurrence. Despite these advancements, the clinical integration of AI in AF management remains an evolving field. Future research should focus on large-scale validation, model interpretability, and regulatory frameworks to ensure widespread adoption. This review explores the current and emerging applications of AI in AF, highlighting its potential to enhance precision medicine and patient outcomes.

Assessing the validity of Society of Thoracic Surgeons (STS) score in predicting stroke risk among patients undergoing cardiothoracic surgery at a tertiary hospital in Pakistan: a retrospective cohort study

BackgroundPostoperative stroke is one of the most serious consequences of cardiac surgery. Morbidity risk assessment is critical for preoperative risk assessments and resource allocation. In this article, we aim to investigate the predictive value of Society of Thoracic Surgeons (STS) score’s effectiveness in stroke risk in cardiothoracic surgery patients in our population.MethodsThis retrospective cohort study was conducted at Aga Khan University Hospital (AKUH) using a consecutive sampling technique. The study included all eligible patients aged 18 years or older who underwent cardiac surgical procedures between January 2010 and December 2016. Of the 3,898 patients initially identified, 814 records were excluded due to incomplete data or pre-existing conditions. Statistical analyses, including chi-square tests, t-tests, and logistic regression, were performed to identify significant predictors of stroke. Prediction accuracy was assessed using a Receiver Operating Characteristic (ROC) curve, with Youden’s J statistic employed to determine optimal sensitivity and specificity thresholds.ResultsOut of 3,084 patients, 52 (1.7%) experienced a postoperative stroke. Stroke patients were significantly older (mean age 62.8 years vs. 57.7 years). They also had higher white blood cell count (10.7 ± 5.1 vs. 9.3 ± 3.3) and a longer history of myocardial infarction (9.2 ± 9.3 years vs. 6.9 ± 7.6 years). Cardiovascular interventions (15.4% vs. 7.7%), postoperative congestive heart failure (21.1% vs. 7.3%), and use of inotropes (5.8% vs. 1.6%) were more prevalent in stroke patients. Emergent surgical status (19.2% vs. 13.4%) and complications such as dialysis, prolonged ventilation, and intra-aortic balloon pump use were also significantly higher. The predictive model demonstrated strong accuracy in predicting postoperative stroke (AUC: 0.841, CI: 0.794–0.888). The ROC analysis for the STS stroke model showed high sensitivity (90.4%) and negative predictive value (99.7%), with moderate specificity (64.3%) and overall accuracy (64.8%), indicating excellent performance in ruling out stroke but moderate reliability in identifying positive cases.ConclusionThe STS risk score demonstrated strong predictive accuracy for postoperative stroke risk in cardiothoracic surgery patients, effectively incorporating clinical factors already accounted for in the comprehensive set of 70 variables used in its calculation.

Enhancing neurological disease diagnostics: fusion of deep transfer learning with optimization algorithm for acute brain stroke prediction using facial images

Stroke is a main risk to life and fitness in current society, particularly in the aging population. Also, the stroke is recognized as a cerebrovascular accident. It contains a nervous illness, which can result from haemorrhage or ischemia of the brain veins, and regular mains to assorted motor and cognitive damages that cooperate with functionality. Screening for stroke comprises physical examination, history taking, and valuation of risk features like age or certain cardiovascular illnesses. Symptoms and signs of stroke include facial weakness. Even though computed tomography (CT) and magnetic resonance imaging (MRI) are standard diagnosis techniques, artificial intelligence (AI) systems have been constructed based on these methods, which deliver fast detection. AI is gaining high attention and is being combined into numerous areas with medicine to enhance the accuracy of analysis and the quality of patient care. This paper proposes an enhancing neurological disease diagnostics fusion of transfer learning for acute brain stroke prediction using facial images (ENDDFTL-ABSPFI) method. The proposed ENDDFTL-ABSPFI method aims to enhance brain stroke detection and classification models using facial imaging. Initially, the image pre-processing stage applies the fuzzy-based median filter (FMF) model to eliminate the noise in input image data. Furthermore, fusion models such as Inception-V3 and EfficientNet-B0 perform the feature extraction. Moreover, the hybrid of convolutional neural network and bidirectional long short-term memory (CNN-BiLSTM) model is employed for the brain stroke classification process. Finally, the multi‐objective sailfish optimization (MOSFO)-based hyperparameter selection process is carried out to optimize the classification outcomes of the CNN-BiLSTM model. The simulation validation of the ENDDFTL-ABSPFI technique is investigated under the Kaggle dataset concerning various measures. The comparative evaluation of the ENDDFTL-ABSPFI technique portrayed a superior accuracy value of 98.60% over existing methods.

Clinical and Neuroimaging Predictors of Stroke in Sickle Cell Disease Patients Who Underwent Stem Cell Transplantation: A Multi-Center Cross Sectional Study. (S22.004)

Clinical and Neuroimaging Predictors of Stroke in Sickle Cell Disease Patients Who Underwent Stem Cell Transplantation: A Multi-Center Cross Sectional Study. (S22.004)

Predicting the Occurrence of Cerebrovascular Accident in Patients using Machine Learning Technique

Cerebrovascular disease commonly known as stroke is the third leading cause of disability and mortality in the world. In recent years, technological advancements have transformed the way information is acquired and how problems are solved in diverse fields of human endeavors, including the medical and healthcare sectors. Machine Learning (ML) and data driven techniques have gain prominence in problem solving and have been deployed in the prediction of the occurrences of stroke. This work explores the application of supervised machine learning algorithms for the prediction of stroke, emphasizing the critical need for early prediction to enhance preventive measures. A comprehensive comparison of classification (Support Vector Machine and Random Forest) and regression (Logistic Regression) algorithms was conducted, with concerns on binary stroke outcome (likelihood of stroke and no stroke) data utilizing dataset from the International Stroke Trial database. The Synthetic Minority Oversampling Technique (SMOTE) and K-fold cross validation were used to balance and address the class imbalance in the datasets. The subsequent model comparison demonstrated distinct strengths and weaknesses among the three models. Random Forest (RF) exhibited high accuracy score of 89%, Support Vector Machine (SVM) and Logistic Regression (LR) showed 86% accuracy. LR demonstrated the most balanced predictive performance, achieving high precision for stroke cases and reasonable recall for both classes.

AI Powered Stroke Prediction and Prevention System Using NLP

This system offers an advanced solution for stroke prediction by combining Artificial Intelligence (AI) and Natural Language Processing (NLP) to enhance early detection and prevention. It analyses structured health data, including medical records and wearable vitals, alongside unstructured data from clinical notes, patient reports, and online health forums. Advanced machine learning techniques such as Random Forest and Gradient Boosting are applied for accurate stroke risk prediction. NLP methods like Named Entity Recognition (NER) and sentiment analysis extract valuable insights from unstructured medical text. The model delivers personalized prevention strategies, including lifestyle changes, dietary recommendations, and early medical intervention. Real-time alert mechanisms are also integrated to detect warning signs and initiate emergency responses. This system helps prevent strokes by enabling timely risk detection, improving patient care, and supporting early, personalized treatment decisions using AI-driven insights. Key Words: Stroke prediction, Artificial Intelligence (AI), Natural Language Processing (NLP), Early detection, Random Forest, Gradient Boosting, Named Entity Recognition (NER), Personalized prevention.

Analyzing the Impact of Data Resampling on Stroke Prediction using Machine Learning

This study focuses on stroke prediction using machine learning algorithms and evaluates the impact of different resampling techniques, including original, under-sampling, and over-sampling, on classification performance. The classifiers used in this study include Random Forest (RF), Decision Tree (DT), Gradient Boosting (GB), and K-Nearest Neighbor (KNN). Each model was trained and evaluated using performance metrics such as accuracy, precision, recall, F1-score, and AUC. The results demonstrate that RF trained on the oversampled dataset achieved the best performance with an accuracy of 94.31%, a precision of 93.52%, a recall of 95.27%, an F1-score of 94.39%, and an AUC of 98.46% on the test set. These findings highlight the effectiveness of oversampling in handling imbalanced datasets and the superiority of RF in stroke prediction tasks compared to other classification methods and resampling techniques.

Independent predictors of ischemic stroke: a comprehensive scan of the internal carotid and vertebrobasilar systems.

Stroke has the highest individual disability rate of all ischemic cardiovascular diseases. Thus, the early identification of its risk factors are crucial to reduce its incidence and improve unfavorable prognoses. This study investigated the value of the most reliable predictors of ischemic stroke by comprehensively scanning the internal carotid artery system and vertebrobasilar artery system using integrated cervicocerebral ultrasound (ICCUS). Patients who had undergone ICCUS scanning and who had recently had a non-cardiac ischemic stroke (<30 days) were prospectively included in this study. Relevant ICCUS parameters were recorded for the carotid, vertebral, subclavian, and intracranial arteries, and intracranial collaterals. The independent predictors of ischemic stroke were examined by multivariate logistic regression analysis. The receiver operating characteristic curves and area under the curve (AUC) were used to evaluate the predictive effects. From July 2023 to April 2024, 127 patients were included in the final analysis, of whom, 42 had suffered a stroke and 85 had not suffered a stroke. The multivariate logistic regression analysis showed that the presence of intraplaque hemorrhage (IPH), multiple stenoses in the anterior and posterior cerebral arteries, and opening of the anterior communicating artery within the cerebral collateral on ICCUS scanning were independent predictors of ischemic stroke (all P<0.05). The corresponding odds ratios for these factors were 5.20, 21.63, 19.87, and 4.58, respectively, and the AUC was 0.802 (95% confidence interval: 0.71-0.89, P<0.001). In addition, if the IPH predictor for cervical vessels was combined with any of the one to three predictors for the intracranial vessels, the combined the diagnostic accuracies progressively increased to 68.1-71.4%, 74.5-76.0%, and 80.2%, respectively. This study used ICCUS, a comprehensive, accurate, and efficient ultrasound technology, to identify more reliable independent predictors of ischemic stroke. Our findings may guide the early prevention and timely treatment of patients.

Markers of Left Atrial Myopathy: Prognostic Usefulness for Ischemic Stroke and Dementia in People in Sinus Rhythm.

Various measures of abnormal left atrial (LA) structure or function (LA myopathy) are associated with a higher risk of ischemic stroke and dementia, independent of atrial fibrillation. However, limited data exist on their prognostic usefulness. Therefore, we aimed to assess the ability of markers of LA myopathy to improve the prediction of ischemic stroke and dementia. The ARIC study (Atherosclerosis Risk in Communities) is a prospective community-based cohort study. For this analysis, we included participants who attended visit 5 (2011-2013) without a history of stroke or atrial fibrillation and had a 12-lead ECG and a transthoracic echocardiogram. Markers of LA myopathy included P wave abnormalities from 12-lead ECG, NT-proBNP (N-terminal pro-B-type natriuretic peptide), and LA volume and strain parameters from the echocardiogram. The primary composite outcome comprised ischemic stroke and dementia, which were ascertained through hospital surveillance, cohort follow-up, and death registries. To determine improvement in risk prediction of the composite outcome, each marker was individually added to a model that included CHA2DS2-VASc variables, and Akaike information criterion, C statistic, and its change were computed. Cox proportional hazards models were used to assess the independent association of LA myopathy markers with the outcome. Among 4712 participants (59% female; mean age, 74 years), 193 ischemic strokes and 769 dementia cases were ascertained over a median follow-up of 8.3 years. Of LA myopathy markers, only LA reservoir strain and NT-proBNP significantly improved C statistic when added to the CHA2DS2-VASc model (base C statistic, 0.677) for the prediction of the composite outcome. Adding the LA reservoir yielded the highest increase in C statistic (0.010 [95% CI, 0.003-0.017]), and the model including the LA reservoir showed the lowest Akaike information criterion. In multivariable regression models, LA volume index, NT-proBNP, and LA strain parameters were significantly associated with the composite outcome. Of various LA myopathy markers, LA reservoir yields the greatest improvement in the prediction of ischemic stroke and dementia, supporting its use to identify people at high risk of cerebrovascular events and dementia.

Predictors of Postoperative Stroke After Transfemoral Carotid Artery Stenting

Predictors of Postoperative Stroke After Transfemoral Carotid Artery Stenting

Predictors for recurrent ischemic stroke in patients with watershed infarct caused by intracranial artery stenosis

Predictors for recurrent ischemic stroke in patients with watershed infarct caused by intracranial artery stenosis

Cardiometabolic index predicts cardiovascular events in aging population: a machine learning-based risk prediction framework from a large-scale longitudinal study.

While the Cardiometabolic Index (CMI) serves as a novel marker for assessing adipose tissue distribution and metabolic function, its prognostic utility for cardiovascular disease (CVD) events remains incompletely understood. This investigation sought to elucidate the predictive capabilities of CMI for cardiovascular outcomes and explore underlying mechanistic pathways to establish a comprehensive risk prediction framework. The study encompassed 7,822 individuals from a national health and retirement longitudinal cohort, with participants stratified by CMI quartiles. Following baseline characteristic comparisons and CVD incidence rate calculations, we implemented multiple Cox regression models to assess CMI's cardiovascular risk prediction capabilities. For nomogram construction, we utilized an ensemble machine learning framework, combining Boruta algorithm-based feature selection with Random Forest (RF) and XGBoost analyses to determine key predictive parameters. Throughout the median follow-up duration of 84 months, we documented 1,500 incident CVD cases, comprising 1,148 cardiac events and 488 cerebrovascular events. CVD incidence demonstrated a positive gradient across ascending CMI quartiles. Multivariate Cox regression analysis, adjusting for potential confounders, confirmed a significant association between CMI and CVD risk. Notably, mediation analyses revealed that hypertension and glycated hemoglobin (HbA1c) potentially serve as mechanistic intermediaries in the CMI-CVD relationship. Sex-stratified analyses suggested differential predictive patterns between gender subgroups. Given CMI's robust and consistent predictive capability for stroke outcomes, we developed a machine learning-derived nomogram incorporating five key predictors: age, CMI, hypertension status, high-sensitivity C-reactive protein (hsCRP) and renal function (measured as serum creatinine). The nomogram demonstrated strong discriminative ability, achieving areas under the receiver operating characteristic curve (AUC) of 0.76 (95% CI: 0.56-0.97) and 0.74 (95% CI: 0.66-0.81) for 2-year and 6-year stroke prediction, respectively. Our findings establish CMI as a significant predictor of cardiovascular events in the aging population, with the relationship partially mediated through hypertension and insulin resistance pathways. The validated nomogram, developed using longitudinal data from a substantial elderly cohort, incorporates CMI to enable preclinical risk stratification, supporting timely preventive strategies.

Analysis of Grip Strength Thresholds for Stroke Management and Prevention in South Korean Older Adults.

Muscle weakness in older adults is associated with cardiovascular disease and all-cause mortality. However, its association with stroke prevalence remains underexplored. This study aimed to analyze the absolute grip strength (AGS) and weight-adjusted relative grip strength (RGS) thresholds for stroke prediction in South Korean older adults and to assess their sex-specific predictive ability. Data from 5185 older adults (2231 men; 2954 women) from the Korea National Health and Nutrition Examination Survey (KNHNES, 2014-2018) were analyzed using complex sampling methods. Receiver operating characteristic (ROC) curve analysis was performed to determine AGS and RGS thresholds and predictive performance, while multivariate logistic regression was used to adjust for confounders and to assess independent effects. In older men, both the AGS and RGS demonstrated significant predictive ability for stroke, with AUCs of 0.637 and 0.623, respectively. In women, the AGS (AUC: 0.608) and RGS (AUC: 0.615) were predictive; however, only the RGS was significant for stroke management (odds ratio (OR): 3.026; 95% confidence interval (CI), 1.541-5.943). In men, AGS (OR: 3.544, 95% CI, 2.094-5.998) and RGS (OR: 2.585, 95% CI, 1.529-4.369) were significant. The stroke prediction thresholds were AGS 28.55 kg and RGS 0.47 for men and RGS 0.36 for women. The AGS and RGS provide practical indicators for stroke risk prediction based on sex-specific differences, highlighting their potential for public health and clinical applications. Future studies should investigate the stroke type, severity, and additional functional fitness indices.

A graph neural network technique for the prediction of cerebral stroke using an unbalanced medical dataset

A graph neural network technique for the prediction of cerebral stroke using an unbalanced medical dataset

Recurrent Stroke Prediction by Applying a Stroke Polygenic Risk Score in the Japanese Population.

Recently, various polygenic risk score (PRS)-based methods were developed to improve stroke prediction. However, current PRSs (including cross-ancestry PRS) poorly predict recurrent stroke. Here, we aimed to determine whether the best PRS for Japanese individuals can also predict stroke recurrence in this population by extensively comparing the methods and maximizing the predictive performance for stroke onset. We used data from the disease-oriented BBJ1 (BioBank Japan first cohort; recruited between 2003 and 2007, n=179 938) to derive and optimize the PRSs using a 10-fold cross-validation. We integrated the optimized PRSs for multiple traits, such as vascular risk factors and stroke subtypes to generate a single PRS using the meta-scoring approach (metaGRS). We used an independent BBJ2 (BBJ second cohort; recruited between 2012 and 2017, n=41 929) as a test sample to evaluate the association of the metaGRS with stroke and recurrent stroke. In addition, we analyzed its association stratified by risk factors. We administered 3 distinct tests to consider the potential index event bias. We analyzed recurrent stroke cases (n=174) and nonrecurrent stroke controls (n=1153) among subjects within the BBJ2. After adjusting for known risk factors, metaGRS was associated with stroke recurrence (adjusted odds ratio per SD, 1.18 [95% CI, 1.00-1.39]; P=0.044), although no significant correlation was observed with the published PRSs. The outcomes derived from these examinations did not provide any significant indication of the influence of index event bias. The high metaGRS group without a history of hypertension had a higher risk of stroke recurrence than that of the low metaGRS group (adjusted odds ratio, 2.24 [95% CI, 1.07-4.66]; P=0.032). There was no association at all in the hypertension group (adjusted odds ratio, 1.21 [95% CI, 0.69-2.13]; P=0.50). The metaGRS developed in a Japanese cohort predicted stroke recurrence in an independent cohort of patients. In particular, it predicted an increased risk of recurrence among stroke patients without hypertension. These findings provide clues for additional genetic risk stratification and help in developing personalized strategies for stroke recurrence prevention.

Cross-sectional study on smoking types and stroke risk: development of a predictive model for identifying stroke risk.

Stroke, a major global health concern, is responsible for high mortality and long-term disabilities. With the aging population and increasing prevalence of risk factors, its incidence is on the rise. Existing risk assessment tools have limitations, and there is a pressing need for more accurate and personalized stroke risk prediction models. Smoking, a significant modifiable risk factor, has not been comprehensively examined in current models regarding different smoking types. Data were sourced from the 2015-2018 National Health and Nutrition Examination Survey (NHANES) and the 2020-2021 Behavioral Risk Factor Surveillance System (BRFSS). Tobacco use (including combustible cigarettes and e-cigarettes) and stroke history were obtained through questionnaires. Participants were divided into four subgroups: non-smokers, exclusive combustible cigarette users, exclusive e-cigarette users, and dual users. Covariates such as age, sex, race, education, and health conditions were also collected. Multivariate logistic regression was used to analyze the relationship between smoking and stroke. Four machine-learning models (XGBoost, logistic regression, Random Forest, and Gaussian Naive Bayes) were evaluated using the area under the receiver-operating characteristic curve (AUC), and Shapley's additive interpretation method was applied for feature importance ranking and model interpretation. A total of 273,028 individuals were included in the study. Exclusive combustible cigarette users had an elevated stroke risk (β: 1.36, 95% CI: 1.26-1.47, P < 0.0001). Among the four machine-learning models, the XGBoost model showed the best discriminative ability with an AUC of 0.794 (95% CI = 0.787-0.802). This study reveals a significant association between smoking types and stroke risk. An XGBoost-based stroke prediction model was established, which has the potential to improve the accuracy of stroke risk assessment and contribute to personalized interventions for stroke prevention, thus alleviating the healthcare burden related to stroke.