Analysis shows machine learning predicts stroke risk using clinical factors in young adults, indicating potential for personalized prevention strategies.
Key Points
KNN and naive Bayes achieved the highest accuracy of 88.9%, enhancing risk prediction for young stroke patients.
Young stroke risk prediction utilized seven classifiers, with features including hypertension and alcohol use being crucial.
Feature importance showed hypertension as a consistent predictor, with dropout loss values revealing key factors affecting outcomes.
Developed machine learning models demonstrate robust performance, suggesting significant potential for early intervention.