Predicting Ischemic Stroke in Patients with Atrial Fibrillation Using Machine Learning
Seonwoo Jung1, Min-Keun Song2, Eunjoo Lee3
1Department of ICT Convergence System Engineering, Chonnam National University, 61186 Gwangju, Republic of Korea.
Insights
This study developed a deep learning model to predict ischemic stroke in atrial fibrillation (AF) patients using Korean health data. The model achieved higher accuracy than existing methods, aiding in stroke prevention strategies.
Area of Science:
- Cardiology
- Neurology
- Medical Informatics
Background:
- Atrial fibrillation (AF) is a significant risk factor for stroke.
- Accurate stroke risk prediction is crucial for preventing initial and recurrent cerebrovascular events.
- Early intervention strategies rely on precise risk stratification in AF patients.
Purpose of the Study:
- To develop a machine learning model for predicting ischemic stroke in patients diagnosed with atrial fibrillation.
- To leverage the extensive Korean National Health Insurance (KNHIS) database for robust risk prediction.
- To identify key features associated with ischemic stroke occurrence in the AF population.
Main Methods:
- Extracted 65-dimensional features from 754,949 AF patients in the KNHIS database.
- Employed logistic regression to identify statistically significant predictors of ischemic stroke.
- Constructed an attention-based deep neural network for ischemic stroke prediction.
Main Results:
- Identified 48 features significantly associated with ischemic stroke (p < 0.001).
- The deep learning model achieved a higher AUROC of 0.727 ± 0.003 compared to CHA2DS2-VASc score (0.651 ± 0.007).
- The model demonstrated superior predictive performance over other machine learning methods in a validation cohort of 150,989 AF patients.
Conclusions:
- The developed deep learning model shows promise for predicting ischemic stroke risk in AF patients.
- This approach can enhance preventive medicine by providing personalized stroke risk scores.
- Identifying associated features aids AF patients in proactive stroke prevention planning.
Background:
Atrial fibrillation (AF) is a well-known risk factor for stroke. Predicting the risk is important to prevent the first and secondary attacks of cerebrovascular diseases by determining early treatment. This study aimed to predict the ischemic stroke in AF patients based on the massive and complex Korean National Health Insurance (KNHIS) data through a machine learning approach.
Methods:
We extracted 65-dimensional features, including demographics, health examination, and medical history information, of 754,949 patients with AF from KNHIS. Logistic regression was used to determine whether the extracted features had a statistically significant association with ischemic stroke occurrence. Then, we constructed the ischemic stroke prediction model using an attention-based deep neural network. The extracted features were used as input, and the occurrence of ischemic stroke after the diagnosis of AF was the output used to train the model.
Results:
We found 48 features significantly associated with ischemic stroke occurrence through regression analysis (p-value < 0.001). When the proposed deep learning model was applied to 150,989 AF patients, it was confirmed that the occurrence ischemic stroke was predicted to be higher AUROC (AUROC = 0.727 ± 0.003) compared to CHA2DS2-VASc score (AUROC = 0.651 ± 0.007) and other machine learning methods.
Conclusions:
As part of preventive medicine, this study could help AF patients prepare for ischemic stroke prevention based on predicted stoke associated features and risk scores.
