Application of Machine Learning Techniques to Identify Data Reliability and Factors Affecting Outcome After Stroke Using Electronic Administrative Records

Machine Learning - Based Bleeding Risk Predictions in Atrial Fibrillation Patients on Direct Oral Anticoagulants

Importance

Accurately predicting major bleeding events in non-valvular atrial fibrillation (AF) patients on direct oral anticoagulants (DOACs) is crucial for personalized treatment and improving patient outcomes, especially with emerging alternatives like left atrial appendage closure devices. The left atrial appendage closure devices reduce stroke risk comparably but with significantly fewer non-procedural bleeding events.

Objective

To evaluate the performance of machine learning (ML) risk models in predicting clinically significant bleeding events requiring hospitalization and hemorrhagic stroke in non-valvular AF patients on DOACs compared to conventional bleeding risk scores (HAS-BLED, ORBIT, and ATRIA) at the index visit to a cardiologist for AF management.

Design

Prognostic modeling with retrospective cohort study design using electronic health record (EHR) data, with clinical follow-up at one-, two-, and five-years.

Setting

University of Pittsburgh Medical Center (UPMC) system.

Participants

24,468 non-valvular AF patients aged ≥18 years treated with DOACs, excluding those with prior history of significant bleeding, other indications for DOACs, on warfarin or contraindicated to DOACs.

Exposures

DOAC therapy for non-valvular AF.

Main Outcomes and Measures

The primary endpoint was clinically significant bleeding requiring hospitalization within one year of index visit. The models incorporated demographic, clinical, and laboratory variables available in the EHR at the index visit.

Results

Among 24,468 patients, 553 (2.3%) had bleeding events within one year, 829 (3.5%) within two years, and 1,292 (5.8%) within five years of index visit. We evaluated multivariate logistic regression and ML models including random forest, classification trees, k-nearest neighbor, naive Bayes, and extreme gradient boosting (XGBoost) which modestly outperformed HAS-BLED, ATRIA, and ORBIT scores in predicting clinically significant bleeding at 1-year follow-up. The best performing model (random forest) showed area under the curve (AUC-ROC) 0.76 (0.70-0.81), G-Mean score of 0.67, net reclassification index 0.14 compared to 0.57 (0.50-0.63), G-Mean score of 0.57 for HASBLED score, p-value for difference <0.001. The ML models had improved performance compared to conventional risk across time-points of 2-year and 5-years and within the subgroup of hemorrhagic stroke. SHAP analysis identified novel risk factors including measures from body mass index, cholesterol profile, and insurance type beyond those used in conventional risk scores.

Conclusions and Relevance

Our findings demonstrate the superior performance of ML models compared to conventional bleeding risk scores and identify novel risk factors highlighting the potential for personalized bleeding risk assessment in AF patients on DOACs.

Explaining predictors of discharge destination assessed along the patients' acute stroke journey

INTRODUCTION

Accurate prediction of outcome destination at an early stage would help manage patients presenting with stroke. This study assessed the predictive ability of three machine learning (ML) algorithms to predict outcomes at four different stages as well as compared the predictive power of stroke scores.

METHODS

Patients presenting with acute stroke to the Canberra Hospital between 2015 and 2019 were selected retrospectively. 16 potential predictors and one target variable (discharge destination) were obtained from the notes. k-Nearest Neighbour (kNN) and two ensemble-based classification algorithms (Adaptive Boosting and Bootstrap Aggregation) were employed to predict outcomes. Predictive accuracy was assessed at each of the four stages using both overall and per-class accuracy. The contribution of each variable to the prediction outcome was evaluated by the ensemble-based algorithm and using the Relief feature selection algorithm. Various combinations of stroke scores were tested using the aforementioned models.

RESULTS

Of the three ML models, Adaptive Boosting demonstrated the highest accuracy (90%) at Stage 4 in predicting death while the highest overall accuracy (81.7%) was achieved by kNN (k=2/City-block distance). Feature importance analysis has shown that the most important features are the 24-hour Scandinavian Stroke Scale (SSS) and 24-hour National Institutes of Health Stroke Scale (NIHSS) scores, dyslipidaemia, hypertension and premorbid mRS score. For the initial and 24-hour scores, there was a higher correlation (0.93) between SSS scores than for NIHSS scores (0.81). Reducing the overall four scores to InitSSS/24hrNIHSS increased accuracy to 95% in predicting death (Adaptive Boosting) and overall accuracy to 85.4% (kNN). Accuracies at Stage 2 (pre-treatment, 11 predictors) were not far behind those at Stage 4.

CONCLUSION

Our findings suggest that even in the early stages of management, a clinically useful prediction regarding discharge destination can be made. Adaptive Boosting might be the best ML model, especially when it comes to predicting death. The predictors' importance analysis also showed that dyslipidemia and hypertension contributed to the discharge outcome even more than expected. Further, surprisingly using mixed score systems might also lead to higher prediction accuracies.

Predicting patient-reported outcome of activities of daily living in stroke rehabilitation: a machine learning study

BACKGROUND

Machine Learning is increasingly used to predict rehabilitation outcomes in stroke in the context of precision rehabilitation and patient-centered care. However, predictors for patient-centered outcome measures for activities and participation in stroke rehabilitation requires further investigation.

METHODS

This study retrospectively analyzed data collected for our previous studies from 124 participants. Machine Learning models were built to predict postintervention improvement of patient-reported outcome measures of daily activities (i.e, the Motor Activity Log and the Nottingham Extended Activities of Daily Living) and participation (i.e, the Activities of Daily Living domain of the Stroke Impact Scale). Three groups of 18 potential predictors were included: patient demographics, stroke characteristics, and baseline assessment scores that encompass all three domains under the framework of International Classification of Functioning, Disability and Health. For each target variable, classification models were built with four algorithms, logistic regression, k-nearest neighbors, support vector machine, and random forest, and with all 18 potential predictors and the most important predictors identified by feature selection.

RESULTS

Predictors for the four target variables partially overlapped. For all target variables, their own baseline scores were among the most important predictors. Upper-limb motor function and selected demographic and stroke characteristics were also among the important predictors across the target variables. For the four target variables, prediction accuracies of the best-performing models with 18 features ranged between 0.72 and 0.96. Those of the best-performing models with fewer features ranged between 0.72 and 0.84.

CONCLUSIONS

Our findings support the feasibility of using Machine Learning for the prediction of stroke rehabilitation outcomes. The study was the first to use Machine Learning to identify important predictors for postintervention improvement on four patient-reported outcome measures of activities and participation in chronic stroke. The study contributes to precision rehabilitation and patient-centered care, and the findings may provide insights into the identification of patients that are likely to benefit from stroke rehabilitation.

Urinary dysfunction in patients with vascular cognitive impairment

Vascular cognitive impairment (VCI) is caused by vascular pathologies, with the spectrum of cognitive disorders ranging from subjective cognitive dysfunction to dementia. Particularly among older adults, cognitive impairment is often complicated with urinary dysfunction (UD); some patients may present with UD before cognitive impairment owing to stroke or even when there are white matter hyperintensities on imaging studies. Patients with cognitive impairment often have both language and movement dysfunction, and thus, UD in patients with VCI can often be underdiagnosed and remain untreated. UD has an impact on the quality of life of patients and caregivers, often leading to poor outcomes. Medical history is an important aspect and should be taken from both patients and their caregivers. Clinical assessment including urinalysis, voiding diary, scales on UD and cognitive impairment, post-void residual volume measurement, uroflowmetry, and (video-) urodynamics should be performed according to indication. Although studies on UD with VCI are few, most of them show that an overactive bladder (OAB) is the most common UD type, and urinary incontinence is the most common symptom. Normal urine storage and micturition in a specific environment are complex processes that require a sophisticated neural network. Although there are many studies on the brain-urinary circuit, the specific circuit involving VCI and UD remains unclear. Currently, there is no disease-modifying pharmacological treatment for cognitive impairment, and anti-acetylcholine drugs, which are commonly used to treat OAB, may cause cognitive impairment, leading to a vicious circle. Therefore, it is important to understand the complex interaction between UD and VCI and formulate individualized treatment plans. This review provides an overview of research advances in clinical features, imaging and pathological characteristics, and treatment options of UD in patients with VCI to increase subject awareness, facilitate research, and improve diagnosis and treatment rates.

Novel machine learning models to predict pneumonia events in supratentorial intracerebral hemorrhage populations: An analysis of the Risa-MIS-ICH study

Background

Stroke-associated pneumonia (SAP) contributes to high mortality rates in spontaneous intracerebral hemorrhage (sICH) populations. Accurate prediction and early intervention of SAP are associated with prognosis. None of the previously developed predictive scoring systems are widely accepted. We aimed to derive and validate novel supervised machine learning (ML) models to predict SAP events in supratentorial sICH populations.

Methods

The data of eligible supratentorial sICH individuals were extracted from the Risa-MIS-ICH database and split into training, internal validation, and external validation datasets. The primary outcome was SAP during hospitalization. Univariate and multivariate analyses were used for variable filtering, and logistic regression (LR), Gaussian naïve Bayes (GNB), random forest (RF), K-nearest neighbor (KNN), support vector machine (SVM), extreme gradient boosting (XGB), and ensemble soft voting model (ESVM) were adopted for ML model derivations. The accuracy, sensitivity, specificity, and area under the curve (AUC) were adopted to evaluate the predictive value of each model with internal/cross-/external validations.

Results

A total of 468 individuals with sICH were included in this work. Six independent variables [nasogastric feeding, airway support, unconscious onset, surgery for external ventricular drainage (EVD), larger sICH volume, and intensive care unit (ICU) stay] for SAP were identified and selected for ML prediction model derivations and validations. The internal and cross-validations revealed the superior and robust performance of the GNB model with the highest AUC value (0.861, 95% CI: 0.793–0.930), while the LR model had the highest AUC value (0.867, 95% CI: 0.812–0.923) in external validation. The ESVM method combining the other six methods had moderate but robust abilities in both cross-validation and external validation and achieved an AUC of 0.843 (95% CI: 0.784–0.902) in external validation.

Conclusion

The ML models could effectively predict SAP in sICH populations, and our novel ensemble model demonstrated reliable robust performance outcomes despite the populational and algorithmic differences. This attempt indicated that ML application may benefit in the early identification of SAP.