Artificial Intelligence-Enabled ECG Algorithm for the Prediction of Coronary Artery Calcification
Coronary artery calcium (CAC), which can be measured in various types of computed tomography (CT) examinations, is a hallmark of coronary artery atherosclerosis. However, despite the clinical value of CAC scores in predicting cardiovascular events, routine measurement of CAC scores is limited due to...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-04-01
|
| Series: | Frontiers in Cardiovascular Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fcvm.2022.849223/full |
| _version_ | 1819024936591687680 |
|---|---|
| author | Changho Han Ki-Woon Kang Tae Young Kim Jae-Sun Uhm Je-Wook Park In Hyun Jung Minkwan Kim SungA Bae Hong-Seok Lim Dukyong Yoon Dukyong Yoon Dukyong Yoon |
| author_facet | Changho Han Ki-Woon Kang Tae Young Kim Jae-Sun Uhm Je-Wook Park In Hyun Jung Minkwan Kim SungA Bae Hong-Seok Lim Dukyong Yoon Dukyong Yoon Dukyong Yoon |
| author_sort | Changho Han |
| collection | DOAJ |
| description | Coronary artery calcium (CAC), which can be measured in various types of computed tomography (CT) examinations, is a hallmark of coronary artery atherosclerosis. However, despite the clinical value of CAC scores in predicting cardiovascular events, routine measurement of CAC scores is limited due to high cost, radiation exposure, and lack of widespread availability. It would be of great clinical significance if CAC could be predicted by electrocardiograms (ECGs), which are cost-effective and routinely performed during various medical checkups. We aimed to develop binary classification artificial intelligence (AI) models that predict CAC using only ECGs as input. Moreover, we aimed to address the generalizability of our model in different environments by externally validating our model on a dataset from a different institution. Among adult patients, standard 12-lead ECGs were extracted if measured within 60 days before or after the CAC scores, and labeled with the corresponding CAC scores. We constructed deep convolutional neural network models based on residual networks using only the raw waveforms of the ECGs as input, predicting CAC at different levels, namely CAC score ≥100, ≥400 and ≥1,000. Our AI models performed well in predicting CAC in the training and internal validation dataset [area under the receiver operating characteristics curve (AUROC) 0.753 ± 0.009, 0.802 ± 0.027, and 0.835 ± 0.024 for the CAC score ≥100, ≥400, and ≥1,000 model, respectively]. Our models also performed well in the external validation dataset (AUROC 0.718, 0.777 and 0.803 for the CAC score ≥100, ≥400, and ≥1,000 model, respectively), indicating that our model can generalize well to different but plausibly related populations. Model performance in terms of AUROC increased in the order of CAC score ≥100, ≥400, and ≥1,000 model, indicating that higher CAC scores might be associated with more prominent structural changes of the heart detected by the model. With our AI models, a substantial proportion of previously unrecognized CAC can be afforded with a risk stratification of CAC, enabling initiation of prophylactic therapy, and reducing the adverse consequences related to ischemic heart disease. |
| first_indexed | 2024-12-21T05:02:42Z |
| format | Article |
| id | doaj.art-1ea06fdee07241ae8b24b6711ae80fda |
| institution | Directory Open Access Journal |
| issn | 2297-055X |
| language | English |
| last_indexed | 2024-12-21T05:02:42Z |
| publishDate | 2022-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Cardiovascular Medicine |
| spelling | doaj.art-1ea06fdee07241ae8b24b6711ae80fda2022-12-21T19:15:11ZengFrontiers Media S.A.Frontiers in Cardiovascular Medicine2297-055X2022-04-01910.3389/fcvm.2022.849223849223Artificial Intelligence-Enabled ECG Algorithm for the Prediction of Coronary Artery CalcificationChangho Han0Ki-Woon Kang1Tae Young Kim2Jae-Sun Uhm3Je-Wook Park4In Hyun Jung5Minkwan Kim6SungA Bae7Hong-Seok Lim8Dukyong Yoon9Dukyong Yoon10Dukyong Yoon11Department of Biomedical Systems Informatics, Yonsei University College of Medicine, Yongin, South KoreaDivision of Cardiology, College of Medicine, Heart Research Institute, Chung-Ang University Hospital, Chung-Ang University, Seoul, South KoreaBUD.on Inc., Seoul, South KoreaDivision of Cardiology, Department of Internal Medicine, Yongin Severance Hospital, Yonsei University College of Medicine, Yongin, South KoreaDivision of Cardiology, Department of Internal Medicine, Yongin Severance Hospital, Yonsei University College of Medicine, Yongin, South KoreaDivision of Cardiology, Department of Internal Medicine, Yongin Severance Hospital, Yonsei University College of Medicine, Yongin, South KoreaDivision of Cardiology, Department of Internal Medicine, Yongin Severance Hospital, Yonsei University College of Medicine, Yongin, South KoreaDivision of Cardiology, Department of Internal Medicine, Yongin Severance Hospital, Yonsei University College of Medicine, Yongin, South KoreaDepartment of Cardiology, Ajou University School of Medicine, Suwon, South KoreaDepartment of Biomedical Systems Informatics, Yonsei University College of Medicine, Yongin, South KoreaBUD.on Inc., Seoul, South KoreaCenter for Digital Health, Yongin Severance Hospital, Yonsei University Health System, Yongin, South KoreaCoronary artery calcium (CAC), which can be measured in various types of computed tomography (CT) examinations, is a hallmark of coronary artery atherosclerosis. However, despite the clinical value of CAC scores in predicting cardiovascular events, routine measurement of CAC scores is limited due to high cost, radiation exposure, and lack of widespread availability. It would be of great clinical significance if CAC could be predicted by electrocardiograms (ECGs), which are cost-effective and routinely performed during various medical checkups. We aimed to develop binary classification artificial intelligence (AI) models that predict CAC using only ECGs as input. Moreover, we aimed to address the generalizability of our model in different environments by externally validating our model on a dataset from a different institution. Among adult patients, standard 12-lead ECGs were extracted if measured within 60 days before or after the CAC scores, and labeled with the corresponding CAC scores. We constructed deep convolutional neural network models based on residual networks using only the raw waveforms of the ECGs as input, predicting CAC at different levels, namely CAC score ≥100, ≥400 and ≥1,000. Our AI models performed well in predicting CAC in the training and internal validation dataset [area under the receiver operating characteristics curve (AUROC) 0.753 ± 0.009, 0.802 ± 0.027, and 0.835 ± 0.024 for the CAC score ≥100, ≥400, and ≥1,000 model, respectively]. Our models also performed well in the external validation dataset (AUROC 0.718, 0.777 and 0.803 for the CAC score ≥100, ≥400, and ≥1,000 model, respectively), indicating that our model can generalize well to different but plausibly related populations. Model performance in terms of AUROC increased in the order of CAC score ≥100, ≥400, and ≥1,000 model, indicating that higher CAC scores might be associated with more prominent structural changes of the heart detected by the model. With our AI models, a substantial proportion of previously unrecognized CAC can be afforded with a risk stratification of CAC, enabling initiation of prophylactic therapy, and reducing the adverse consequences related to ischemic heart disease.https://www.frontiersin.org/articles/10.3389/fcvm.2022.849223/fullcoronary artery calciumatherosclerosiscoronary artery diseaseelectrocardiogramartificial intelligencedeep neural network |
| spellingShingle | Changho Han Ki-Woon Kang Tae Young Kim Jae-Sun Uhm Je-Wook Park In Hyun Jung Minkwan Kim SungA Bae Hong-Seok Lim Dukyong Yoon Dukyong Yoon Dukyong Yoon Artificial Intelligence-Enabled ECG Algorithm for the Prediction of Coronary Artery Calcification Frontiers in Cardiovascular Medicine coronary artery calcium atherosclerosis coronary artery disease electrocardiogram artificial intelligence deep neural network |
| title | Artificial Intelligence-Enabled ECG Algorithm for the Prediction of Coronary Artery Calcification |
| title_full | Artificial Intelligence-Enabled ECG Algorithm for the Prediction of Coronary Artery Calcification |
| title_fullStr | Artificial Intelligence-Enabled ECG Algorithm for the Prediction of Coronary Artery Calcification |
| title_full_unstemmed | Artificial Intelligence-Enabled ECG Algorithm for the Prediction of Coronary Artery Calcification |
| title_short | Artificial Intelligence-Enabled ECG Algorithm for the Prediction of Coronary Artery Calcification |
| title_sort | artificial intelligence enabled ecg algorithm for the prediction of coronary artery calcification |
| topic | coronary artery calcium atherosclerosis coronary artery disease electrocardiogram artificial intelligence deep neural network |
| url | https://www.frontiersin.org/articles/10.3389/fcvm.2022.849223/full |
| work_keys_str_mv | AT changhohan artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification AT kiwoonkang artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification AT taeyoungkim artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification AT jaesunuhm artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification AT jewookpark artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification AT inhyunjung artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification AT minkwankim artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification AT sungabae artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification AT hongseoklim artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification AT dukyongyoon artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification AT dukyongyoon artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification AT dukyongyoon artificialintelligenceenabledecgalgorithmforthepredictionofcoronaryarterycalcification |