Artificial intelligent-based personalized predictive ischemic stroke among type 2 diabetes mellitus complication patients

Objectives The objective of this study was to apply machine learning algorithms to predict the risk of ischemic stroke in type 2 diabetes mellitus patients who were prescribed antidiabetic medications. This is an important complication of type 2 diabetes, and developing prediction models can help identify patients at a higher risk of developing it. Method The study used a dataset of 39,646 patients with type 2 diabetes from the Taipei Medical University Clinical Research Database between 2008 and 2020. The performance of different machine-learning models was evaluated using several metrics, such as the area under the curve, sensitivity, specificity, F1-score, and others. Results The results showed promising outcomes, with the area under the curve improving from 0.67 to 0.78. The critical factors in the machine learning models were age, stroke history, and antithrombotic medication. Conclusions The development of machine learning algorithms to predict the risk of ischemic stroke in type 2 diabetes patients is a significant contribution to the field. Healthcare providers can use this information to take preventative measures and reduce the risk of stroke in high-risk patients. However, further exploration is required to ensure the accuracy and applicability of the model to other patient populations.