A machine learning-based model for predicting paroxysmal and persistent atrial fibrillation based on EHR
Abstract Background There is no effective way to accurately predict paroxysmal and persistent atrial fibrillation (AF) subtypes unless electrocardiogram (ECG) observation is obtained. We aim to develop a predictive model using a machine learning algorithm for identification of paroxysmal and persist...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-02-01
|
| Series: | BMC Medical Informatics and Decision Making |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12911-025-02880-5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823861966241792000 |
|---|---|
| author | Yuqi Zhang Sijin Li Peibiao Mai Yanqi Yang Niansang Luo Chao Tong Kuan Zeng Kun Zhang |
| author_facet | Yuqi Zhang Sijin Li Peibiao Mai Yanqi Yang Niansang Luo Chao Tong Kuan Zeng Kun Zhang |
| author_sort | Yuqi Zhang |
| collection | DOAJ |
| description | Abstract Background There is no effective way to accurately predict paroxysmal and persistent atrial fibrillation (AF) subtypes unless electrocardiogram (ECG) observation is obtained. We aim to develop a predictive model using a machine learning algorithm for identification of paroxysmal and persistent AF, and investigate the influencing factors. Methods We collected demographic data, medication use, serological indicators, and baseline cardiac ultrasound data of all included subjects, totaling 50 variables. The diagnosis of AF subtypes is confirmed by ECG observation for at least more than 7 days. Variable selection was performed by spearman correlation analysis, recursive feature elimination, and least absolute shrinkage and selection operator regression. We built a prediction model for AF using three machine learning methods. Finally, the significance of each variable was analyzed by Shapley additive explanations method. Results After screening, we found the optimal variable set consisting of 10 variables. The model we built achieved good predictive performance (AUC = 0.870, 95%CI 0.858 to 0.882), and had specificity of 0.851 (95%CI 0.844 to 0.858) and sensitivity of 0.716 (95%CI 0.676 to 0.755). Good predictive performance was stably achieved in different age subgroups and different gender subgroups. LA and NT-proBNP were the two most important variables for predicting paroxysmal and persistent AF in all models, except for the female subgroup aged less than 60 years. Conclusions Our model makes it possible to predict paroxysmal and persistent AF based on baseline data at admission. Early and individualized intervention strategies based on our model may help to improve clinical outcomes in AF patients. |
| format | Article |
| id | doaj-art-65d8f81495b948c3904e9082a075979b |
| institution | Kabale University |
| issn | 1472-6947 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Informatics and Decision Making |
| spelling | doaj-art-65d8f81495b948c3904e9082a075979b2025-02-09T12:40:18ZengBMCBMC Medical Informatics and Decision Making1472-69472025-02-0125111510.1186/s12911-025-02880-5A machine learning-based model for predicting paroxysmal and persistent atrial fibrillation based on EHRYuqi Zhang0Sijin Li1Peibiao Mai2Yanqi Yang3Niansang Luo4Chao Tong5Kuan Zeng6Kun Zhang7School of Computer Science & Engineering, Beihang UniversityDepartment of Cardiology, Joint Laboratory of Guangdong-Hong Kong-Macao Universities for Nutritional Metabolism and Precise Prevention and Control of Major Chronic Diseases, The Eighth Affiliated Hospital, Sun Yat-Sen UniversityDepartment of Cardiology, Fuwai Hospital, Chinese Academy of Medical Sciences (Shenzhen Sun Yat-Sen Cardiovascular Hospital)Department of Cardiovascular Surgery, The Eighth Affiliated Hospital, Sun Yat-Sen UniversityDepartment of Cardiology, Sun Yat-Sen Memorial Hospital of Sun Yat-Sen UniversitySchool of Computer Science & Engineering, Beihang UniversityDepartment of Cardiovascular Surgery, The Eighth Affiliated Hospital, Sun Yat-Sen UniversityDepartment of Cardiology, Sun Yat-Sen Memorial Hospital of Sun Yat-Sen UniversityAbstract Background There is no effective way to accurately predict paroxysmal and persistent atrial fibrillation (AF) subtypes unless electrocardiogram (ECG) observation is obtained. We aim to develop a predictive model using a machine learning algorithm for identification of paroxysmal and persistent AF, and investigate the influencing factors. Methods We collected demographic data, medication use, serological indicators, and baseline cardiac ultrasound data of all included subjects, totaling 50 variables. The diagnosis of AF subtypes is confirmed by ECG observation for at least more than 7 days. Variable selection was performed by spearman correlation analysis, recursive feature elimination, and least absolute shrinkage and selection operator regression. We built a prediction model for AF using three machine learning methods. Finally, the significance of each variable was analyzed by Shapley additive explanations method. Results After screening, we found the optimal variable set consisting of 10 variables. The model we built achieved good predictive performance (AUC = 0.870, 95%CI 0.858 to 0.882), and had specificity of 0.851 (95%CI 0.844 to 0.858) and sensitivity of 0.716 (95%CI 0.676 to 0.755). Good predictive performance was stably achieved in different age subgroups and different gender subgroups. LA and NT-proBNP were the two most important variables for predicting paroxysmal and persistent AF in all models, except for the female subgroup aged less than 60 years. Conclusions Our model makes it possible to predict paroxysmal and persistent AF based on baseline data at admission. Early and individualized intervention strategies based on our model may help to improve clinical outcomes in AF patients.https://doi.org/10.1186/s12911-025-02880-5Atrial fibrillationParoxysmal atrial fibrillationPersistent atrial fibrillationMachine learningPrediction model |
| spellingShingle | Yuqi Zhang Sijin Li Peibiao Mai Yanqi Yang Niansang Luo Chao Tong Kuan Zeng Kun Zhang A machine learning-based model for predicting paroxysmal and persistent atrial fibrillation based on EHR BMC Medical Informatics and Decision Making Atrial fibrillation Paroxysmal atrial fibrillation Persistent atrial fibrillation Machine learning Prediction model |
| title | A machine learning-based model for predicting paroxysmal and persistent atrial fibrillation based on EHR |
| title_full | A machine learning-based model for predicting paroxysmal and persistent atrial fibrillation based on EHR |
| title_fullStr | A machine learning-based model for predicting paroxysmal and persistent atrial fibrillation based on EHR |
| title_full_unstemmed | A machine learning-based model for predicting paroxysmal and persistent atrial fibrillation based on EHR |
| title_short | A machine learning-based model for predicting paroxysmal and persistent atrial fibrillation based on EHR |
| title_sort | machine learning based model for predicting paroxysmal and persistent atrial fibrillation based on ehr |
| topic | Atrial fibrillation Paroxysmal atrial fibrillation Persistent atrial fibrillation Machine learning Prediction model |
| url | https://doi.org/10.1186/s12911-025-02880-5 |
| work_keys_str_mv | AT yuqizhang amachinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT sijinli amachinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT peibiaomai amachinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT yanqiyang amachinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT niansangluo amachinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT chaotong amachinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT kuanzeng amachinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT kunzhang amachinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT yuqizhang machinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT sijinli machinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT peibiaomai machinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT yanqiyang machinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT niansangluo machinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT chaotong machinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT kuanzeng machinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr AT kunzhang machinelearningbasedmodelforpredictingparoxysmalandpersistentatrialfibrillationbasedonehr |