Search

Background: Stroke patients are at high risk of developing cognitive impairment and dementia. Failure to identify cognitive impairment in time could hasten the progression of dementia and impact the rehabilitation plan. Therefore, a reliable method is needed to determine a stroke survivor’s susceptibility to post-stroke cognitive impairment and dementia (PSCID). Method: We conducted a retrospective cohort study of cryptogenic stroke (CS) patients from January 1, 2017, to February 28, 2022, using EHR data to query for PSCID onset. A machine learning (ML) model was created to forecast the occurrence of PSCID during follow-up. The features used in the model included elements from EHR that were found to be associated with PSCID in other studies, including sociodemographic, medical, and mental morbidities. Logistic Regression, Random Forest (RF) Classifier, and Gradient Boosting are examples of ML algorithms that were used. The final model used RF Classifier due to its superior performance. Results: Of 390 CS patients (62±16 years, 56.4% male) included in the analysis, 110 (28.2%) had documented PSCID in EHR following the initial stroke. We evaluated our model in a repeated (n=100) 10-fold cross validation scheme and used Synthetic Minority Oversampling Technique (SMOTE) to compensate for class imbalance. We identified the most informative ten features using an Extra Tree classifier, which are age, transient ischemic attack, sex, smoking, diabetes, hypertension, atherosclerosis, anemia, and atrial fibrillation. Using these features, our RF classifier reached a performance of 71.2±6.5% accuracy, 70±4.9% precision, 75±12% recall, and 0.777±0.082 AUC. Conclusion: Our model could predict the CS patients at risk for PSCID with reasonable accuracy using only ten features. Future work should involve a larger cohort along with more advanced machine learning algorithms to enhance the prediction performance.

Introduction: As stroke survivors transition from acute to post-acute care, and finally to community settings, the Centers for Disease Control reports ~65% receive NO rehabilitation. Even more receive rehabilitation too late, after critical brain changes for recovery are complete. Stroke survivors with invisible disabilities of cognition and depression are especially vulnerable to experience poor recovery. We launched a dedicated process to identify invisible disabilities. Our long-term objective is to make acute and post-acute, evidence-based intervention accessible. Hypothesis: >50% of acute stroke patients have cognitive deficits, or depression. Methods: Our comprehensive stroke center completes bedside psychometric assessment with standardized instruments for aphasia (Language Screening Test, LAST), spatial neglect (Catherine Bergego Scale, CBS), memory/global cognition (Montreal Cognitive Assessment, MoCA), delirium (3-Minute Diagnostic Interview for the Confusion Assessment Method, 3D-CAM) and depression (Patient health questionnaire, PHQ-8). Patients unable to respond to questions are assessed for spatial neglect and delirium (standardized observations). Results: 105 ischemic stroke survivors were assessed in the first quarter of program launch (April-July, 2021). Of that group, patients met screening criteria for spatial neglect (47%), aphasia (40%), delirium (19%) and depression (31%). Over 90% had memory / global cognitive impairment (MoCA Conclusions: Our initiative, which includes systematic acute stroke unit spatial neglect screening, confirmed the previously reported high rate of cognitive disorders and depression (Champod, Eskes, Barrett, 2020). Our current step implements uniform recommendations for patients with deficits, and will examine post-acute outcomes, number receiving rehabilitation and medical follow-up, and treatment disparities (right/left stroke, under-represented groups).