Electrocardiography-based artificial intelligence predicts the upcoming future of heart failure with mildly reduced ejection fraction
BackgroundHeart failure with mildly reduced ejection fraction (HFmrEF) has emerged as the predominant subtype of heart failure (HF). This study aimed to develop artificial intelligence (AI)-electrocardiography (ECG) to identify and predict the prognosis of patients with HFmrEF.MethodsWe collected 10...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-02-01
|
| Series: | Frontiers in Cardiovascular Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fcvm.2025.1418914/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823861046859792384 |
|---|---|
| author | Dae-Young Kim Sang-Won Lee Dong-Ho Lee Sang-Chul Lee Sang-Chul Lee Ji-Hun Jang Sung-Hee Shin Dae-Hyeok Kim Dae-Hyeok Kim Wonik Choi Wonik Choi Yong-Soo Baek Yong-Soo Baek Yong-Soo Baek |
| author_facet | Dae-Young Kim Sang-Won Lee Dong-Ho Lee Sang-Chul Lee Sang-Chul Lee Ji-Hun Jang Sung-Hee Shin Dae-Hyeok Kim Dae-Hyeok Kim Wonik Choi Wonik Choi Yong-Soo Baek Yong-Soo Baek Yong-Soo Baek |
| author_sort | Dae-Young Kim |
| collection | DOAJ |
| description | BackgroundHeart failure with mildly reduced ejection fraction (HFmrEF) has emerged as the predominant subtype of heart failure (HF). This study aimed to develop artificial intelligence (AI)-electrocardiography (ECG) to identify and predict the prognosis of patients with HFmrEF.MethodsWe collected 104,336 12-lead ECG datasets from April 2009 to December 2021 in a tertiary centre. The AI-ECG encompasses a novel model that combines an automatic labelling preprocessing method with a transformer architecture incorporating a triplet loss for HFmrEF analysis.ResultsThe receiver operating characteristic analyses revealed that the area under the curve of AI-ECG for identifying all types of HF was acceptable [0.873, 95% confidence interval (CI): 0.864–0.893], while that for identifying patients with HFmrEF was relatively lower (0.824, 95% CI: 0.794–0.863) than that for those with HF with reduced ejection fraction (EF) (0.875, 95% CI: 0.844–0.912) and those with normal EF (0.870, 95% CI: 0.842–0.894). The analysis of ECG features showed significant increases in QRS duration (p = 0.001), QT interval (p = 0.045), and corrected QT interval (p = 0.041) with increasing “Severity by Euclidean distance”. Following the predictability analysis with another group of 953 patients for improvements of follow-up EF in HFmrEF, the patients were grouped into three clusters based on the AI-Euclidean distance; Cluster 1 had the most severe cases and poorer outcomes than Clusters 2 (p < 0.001) and 3 (p < 0.001).ConclusionsAI-ECG presents an innovative approach for the prognostic stratification of cardiac contractility in patients with HFmrEF. In patients with HFmrEF, disease progression can be predicted using AI-ECG. |
| format | Article |
| id | doaj-art-9d9957762db64dae9e8c9774f8c1138f |
| institution | Kabale University |
| issn | 2297-055X |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Cardiovascular Medicine |
| spelling | doaj-art-9d9957762db64dae9e8c9774f8c1138f2025-02-10T06:48:34ZengFrontiers Media S.A.Frontiers in Cardiovascular Medicine2297-055X2025-02-011210.3389/fcvm.2025.14189141418914Electrocardiography-based artificial intelligence predicts the upcoming future of heart failure with mildly reduced ejection fractionDae-Young Kim0Sang-Won Lee1Dong-Ho Lee2Sang-Chul Lee3Sang-Chul Lee4Ji-Hun Jang5Sung-Hee Shin6Dae-Hyeok Kim7Dae-Hyeok Kim8Wonik Choi9Wonik Choi10Yong-Soo Baek11Yong-Soo Baek12Yong-Soo Baek13Division of Cardiology, Department of Internal Medicine, Inha University College of Medicine, Incheon, Republic of KoreaDepartment of Electrical and Computer Engineering, Inha University, Incheon, Republic of KoreaDeepCardio Inc., Incheon, Republic of KoreaDeepCardio Inc., Incheon, Republic of KoreaDepartment of Computer Engineering, Inha University, Incheon, Republic of KoreaDivision of Cardiology, Department of Internal Medicine, Inha University College of Medicine, Incheon, Republic of KoreaDivision of Cardiology, Department of Internal Medicine, Inha University College of Medicine, Incheon, Republic of KoreaDivision of Cardiology, Department of Internal Medicine, Inha University College of Medicine, Incheon, Republic of KoreaDeepCardio Inc., Incheon, Republic of KoreaDeepCardio Inc., Incheon, Republic of KoreaDepartment of Information and Communication Engineering, Inha University, Incheon, Republic of KoreaDivision of Cardiology, Department of Internal Medicine, Inha University College of Medicine, Incheon, Republic of KoreaDeepCardio Inc., Incheon, Republic of KoreaSchool of Computer Science, University of Birmingham, Birmingham, United KingdomBackgroundHeart failure with mildly reduced ejection fraction (HFmrEF) has emerged as the predominant subtype of heart failure (HF). This study aimed to develop artificial intelligence (AI)-electrocardiography (ECG) to identify and predict the prognosis of patients with HFmrEF.MethodsWe collected 104,336 12-lead ECG datasets from April 2009 to December 2021 in a tertiary centre. The AI-ECG encompasses a novel model that combines an automatic labelling preprocessing method with a transformer architecture incorporating a triplet loss for HFmrEF analysis.ResultsThe receiver operating characteristic analyses revealed that the area under the curve of AI-ECG for identifying all types of HF was acceptable [0.873, 95% confidence interval (CI): 0.864–0.893], while that for identifying patients with HFmrEF was relatively lower (0.824, 95% CI: 0.794–0.863) than that for those with HF with reduced ejection fraction (EF) (0.875, 95% CI: 0.844–0.912) and those with normal EF (0.870, 95% CI: 0.842–0.894). The analysis of ECG features showed significant increases in QRS duration (p = 0.001), QT interval (p = 0.045), and corrected QT interval (p = 0.041) with increasing “Severity by Euclidean distance”. Following the predictability analysis with another group of 953 patients for improvements of follow-up EF in HFmrEF, the patients were grouped into three clusters based on the AI-Euclidean distance; Cluster 1 had the most severe cases and poorer outcomes than Clusters 2 (p < 0.001) and 3 (p < 0.001).ConclusionsAI-ECG presents an innovative approach for the prognostic stratification of cardiac contractility in patients with HFmrEF. In patients with HFmrEF, disease progression can be predicted using AI-ECG.https://www.frontiersin.org/articles/10.3389/fcvm.2025.1418914/fullartificial intelligenceelectrocardiographyheart failurepredictabilityejection fraction |
| spellingShingle | Dae-Young Kim Sang-Won Lee Dong-Ho Lee Sang-Chul Lee Sang-Chul Lee Ji-Hun Jang Sung-Hee Shin Dae-Hyeok Kim Dae-Hyeok Kim Wonik Choi Wonik Choi Yong-Soo Baek Yong-Soo Baek Yong-Soo Baek Electrocardiography-based artificial intelligence predicts the upcoming future of heart failure with mildly reduced ejection fraction Frontiers in Cardiovascular Medicine artificial intelligence electrocardiography heart failure predictability ejection fraction |
| title | Electrocardiography-based artificial intelligence predicts the upcoming future of heart failure with mildly reduced ejection fraction |
| title_full | Electrocardiography-based artificial intelligence predicts the upcoming future of heart failure with mildly reduced ejection fraction |
| title_fullStr | Electrocardiography-based artificial intelligence predicts the upcoming future of heart failure with mildly reduced ejection fraction |
| title_full_unstemmed | Electrocardiography-based artificial intelligence predicts the upcoming future of heart failure with mildly reduced ejection fraction |
| title_short | Electrocardiography-based artificial intelligence predicts the upcoming future of heart failure with mildly reduced ejection fraction |
| title_sort | electrocardiography based artificial intelligence predicts the upcoming future of heart failure with mildly reduced ejection fraction |
| topic | artificial intelligence electrocardiography heart failure predictability ejection fraction |
| url | https://www.frontiersin.org/articles/10.3389/fcvm.2025.1418914/full |
| work_keys_str_mv | AT daeyoungkim electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT sangwonlee electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT dongholee electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT sangchullee electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT sangchullee electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT jihunjang electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT sungheeshin electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT daehyeokkim electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT daehyeokkim electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT wonikchoi electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT wonikchoi electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT yongsoobaek electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT yongsoobaek electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction AT yongsoobaek electrocardiographybasedartificialintelligencepredictstheupcomingfutureofheartfailurewithmildlyreducedejectionfraction |