1
|
Cho Y, Yoon M, Kim J, Lee JH, Oh IY, Lee CJ, Kang SM, Choi DJ. Artificial Intelligence-Based Electrocardiographic Biomarker for Outcome Prediction in Patients With Acute Heart Failure: Prospective Cohort Study. J Med Internet Res 2024; 26:e52139. [PMID: 38959500 PMCID: PMC11255523 DOI: 10.2196/52139] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/27/2023] [Revised: 02/22/2024] [Accepted: 05/29/2024] [Indexed: 07/05/2024] Open
Abstract
BACKGROUND Although several biomarkers exist for patients with heart failure (HF), their use in routine clinical practice is often constrained by high costs and limited availability. OBJECTIVE We examined the utility of an artificial intelligence (AI) algorithm that analyzes printed electrocardiograms (ECGs) for outcome prediction in patients with acute HF. METHODS We retrospectively analyzed prospectively collected data of patients with acute HF at two tertiary centers in Korea. Baseline ECGs were analyzed using a deep-learning system called Quantitative ECG (QCG), which was trained to detect several urgent clinical conditions, including shock, cardiac arrest, and reduced left ventricular ejection fraction (LVEF). RESULTS Among the 1254 patients enrolled, in-hospital cardiac death occurred in 53 (4.2%) patients, and the QCG score for critical events (QCG-Critical) was significantly higher in these patients than in survivors (mean 0.57, SD 0.23 vs mean 0.29, SD 0.20; P<.001). The QCG-Critical score was an independent predictor of in-hospital cardiac death after adjustment for age, sex, comorbidities, HF etiology/type, atrial fibrillation, and QRS widening (adjusted odds ratio [OR] 1.68, 95% CI 1.47-1.92 per 0.1 increase; P<.001), and remained a significant predictor after additional adjustments for echocardiographic LVEF and N-terminal prohormone of brain natriuretic peptide level (adjusted OR 1.59, 95% CI 1.36-1.87 per 0.1 increase; P<.001). During long-term follow-up, patients with higher QCG-Critical scores (>0.5) had higher mortality rates than those with low QCG-Critical scores (<0.25) (adjusted hazard ratio 2.69, 95% CI 2.14-3.38; P<.001). CONCLUSIONS Predicting outcomes in patients with acute HF using the QCG-Critical score is feasible, indicating that this AI-based ECG score may be a novel biomarker for these patients. TRIAL REGISTRATION ClinicalTrials.gov NCT01389843; https://clinicaltrials.gov/study/NCT01389843.
Collapse
Affiliation(s)
- Youngjin Cho
- Division of Cardiology, Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Gyeonggi-do, Republic of Korea
- ARPI Inc, Seongnam, Gyeonggi-do, Republic of Korea
| | - Minjae Yoon
- Division of Cardiology, Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Gyeonggi-do, Republic of Korea
| | - Joonghee Kim
- ARPI Inc, Seongnam, Gyeonggi-do, Republic of Korea
- Department of Emergency Medicine, Seoul National University Bundang Hospital, Seongnam, Gyeonggi-do, Republic of Korea
| | - Ji Hyun Lee
- Division of Cardiology, Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Gyeonggi-do, Republic of Korea
| | - Il-Young Oh
- Division of Cardiology, Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Gyeonggi-do, Republic of Korea
| | - Chan Joo Lee
- Division of Cardiology, Department of Internal Medicine, Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of Korea
| | - Seok-Min Kang
- Division of Cardiology, Department of Internal Medicine, Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of Korea
| | - Dong-Ju Choi
- Division of Cardiology, Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Gyeonggi-do, Republic of Korea
| |
Collapse
|
2
|
Zhang CJ, Yuan-Lu, Tang FQ, Cai HP, Qian YF, Chao-Wang. Heart failure classification using deep learning to extract spatiotemporal features from ECG. BMC Med Inform Decis Mak 2024; 24:17. [PMID: 38225576 PMCID: PMC10788991 DOI: 10.1186/s12911-024-02415-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/31/2023] [Accepted: 01/03/2024] [Indexed: 01/17/2024] Open
Abstract
BACKGROUND Heart failure is a syndrome with complex clinical manifestations. Due to increasing population aging, heart failure has become a major medical problem worldwide. In this study, we used the MIMIC-III public database to extract the temporal and spatial characteristics of electrocardiogram (ECG) signals from patients with heart failure. METHODS We developed a NYHA functional classification model for heart failure based on a deep learning method. We introduced an integrating attention mechanism based on the CNN-LSTM-SE model, segmenting the ECG signal into 2 to 20 s long segments. Ablation experiments showed that the 12 s ECG signal segments could be used with the proposed deep learning model for superior classification of heart failure. RESULTS The accuracy, positive predictive value, sensitivity, and specificity of the NYHA functional classification method were 99.09, 98.9855, 99.033, and 99.649%, respectively. CONCLUSIONS The comprehensive performance of this model exceeds similar methods and can be used to assist in clinical medical diagnoses.
Collapse
Affiliation(s)
- Chang-Jiang Zhang
- Taizhou Central Hospital, Affiliated Hospital of Taizhou University, Taizhou, China
- School of Electronic and Information Engineering (School of Big Data Science), Taizhou University, Taizhou, China
| | - Yuan-Lu
- School of Electronic and Information Engineering (School of Big Data Science), Taizhou University, Taizhou, China
- College of Physics and Electronic Information Engineering, Zhejiang Normal University, Jinhua, China
| | - Fu-Qin Tang
- Taizhou Central Hospital, Affiliated Hospital of Taizhou University, Taizhou, China.
| | - Hai-Peng Cai
- Taizhou Central Hospital, Affiliated Hospital of Taizhou University, Taizhou, China
| | - Yin-Fen Qian
- Taizhou Central Hospital, Affiliated Hospital of Taizhou University, Taizhou, China
| | - Chao-Wang
- Taizhou Central Hospital, Affiliated Hospital of Taizhou University, Taizhou, China
| |
Collapse
|
3
|
Pokhrel Bhattarai S, Block RC, Xue Y, Rodriguez DH, Tucker RG, Carey MG. Integrative review of electrocardiographic characteristics in patients with reduced, mildly reduced, and preserved heart failure. Heart Lung 2024; 63:142-158. [PMID: 37913557 DOI: 10.1016/j.hrtlng.2023.10.012] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/27/2023] [Revised: 10/09/2023] [Accepted: 10/23/2023] [Indexed: 11/03/2023]
Abstract
INTRODUCTION Electrocardiographic (ECG) changes in heart failure with reduced, mildly reduced, and preserved ejection fractions can be critical in clinical assessment while waiting to perform echocardiograms or when it is unavailable. This integrative review aimed to identify ECG characteristics among hospitalized patients demonstrating three types of heart failure during acute decompensation. METHODS We searched an electronic database of PubMed, Web of Science, EMBASE, Scopus, Google Scholar, and ClinicalTrials.gov using medical subject headings (MeSH) terms and keywords. Sixteen studies were synthesized and reported. RESULTS Heart failure with reduced ejection fraction (HFrEF) was more common in men, comorbid with coronary artery diseases and diabetes mellitus, higher BNP/Pro-BNP, wide QRS, and left bundle branch block on ECG. On average, clients with heart failure with preserved ejection fraction (HFpEF) were older and more likely to have a history of atrial fibrillation, valvular heart diseases, hypertension, chronic obstructive pulmonary, and atrial fibrillation (AF) on ECG. Patients with mildly reduced (HFmrEF) were more similar to HFpEF in older patients, comorbid with hypertension, AF and valvular diseases, and AF on ECG. CONCLUSIONS ECG characteristics might be related to left ventricular ejection fraction. Demographics, BNP/Pro-BNP, and ECG changes might help differentiate different heart failure types. Therefore, ECG might be a prognostic tool while caring for heart failure patients when highly skilled resources are unavailable. These identified ECG characteristics help generate research hypotheses and warrant validation in future research.
Collapse
Affiliation(s)
- Sunita Pokhrel Bhattarai
- University of Rochester School of Nursing, 255 Crittenden Boulevard, Box SON, Rochester, NY 14642, United States.
| | | | - Ying Xue
- University of Rochester School of Nursing, 255 Crittenden Boulevard, Box SON, Rochester, NY 14642, United States
| | - Darcey H Rodriguez
- University of Rochester School of Nursing, 255 Crittenden Boulevard, Box SON, Rochester, NY 14642, United States; University of Rochester Medical Center, United States
| | - Rebecca G Tucker
- University of Rochester School of Nursing, 255 Crittenden Boulevard, Box SON, Rochester, NY 14642, United States
| | - Mary G Carey
- University of Rochester School of Nursing, 255 Crittenden Boulevard, Box SON, Rochester, NY 14642, United States; University of Rochester Medical Center, United States
| |
Collapse
|
4
|
Butler L, Karabayir I, Kitzman DW, Alonso A, Tison GH, Chen LY, Chang PP, Clifford G, Soliman EZ, Akbilgic O. A generalizable electrocardiogram-based artificial intelligence model for 10-year heart failure risk prediction. CARDIOVASCULAR DIGITAL HEALTH JOURNAL 2023; 4:183-190. [PMID: 38222101 PMCID: PMC10787146 DOI: 10.1016/j.cvdhj.2023.11.003] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/16/2024] Open
Abstract
Background Heart failure (HF) is a progressive condition with high global incidence. HF has two main subtypes: HF with preserved ejection fraction (HFpEF) and HF with reduced ejection fraction (HFrEF). There is an inherent need for simple yet effective electrocardiogram (ECG)-based artificial intelligence (AI; ECG-AI) models that can predict HF risk early to allow for risk modification. Objective The main objectives were to validate HF risk prediction models using Multi-Ethnic Study of Atherosclerosis (MESA) data and assess performance on HFpEF and HFrEF classification. Methods There were six models in comparision derived using ARIC data. 1) The ECG-AI model predicting HF risk was developed using raw 12-lead ECGs with a convolutional neural network. The clinical models from 2) ARIC (ARIC-HF) and 3) Framingham Heart Study (FHS-HF) used 9 and 8 variables, respectively. 4) Cox proportional hazards (CPH) model developed using the clinical risk factors in ARIC-HF or FHS-HF. 5) CPH model using the outcome of ECG-AI and the clinical risk factors used in CPH model (ECG-AI-Cox) and 6) A Light Gradient Boosting Machine model using 288 ECG Characteristics (ECG-Chars). All the models were validated on MESA. The performances of these models were evaluated using the area under the receiver operating characteristic curve (AUC) and compared using the DeLong test. Results ECG-AI, ECG-Chars, and ECG-AI-Cox resulted in validation AUCs of 0.77, 0.73, and 0.84, respectively. ARIC-HF and FHS-HF yielded AUCs of 0.76 and 0.74, respectively, and CPH resulted in AUC = 0.78. ECG-AI-Cox outperformed all other models. ECG-AI-Cox provided an AUC of 0.85 for HFrEF and 0.83 for HFpEF. Conclusion ECG-AI using ECGs provides better-validated predictions when compared to HF risk calculators, and the ECG feature model and also works well with HFpEF and HFrEF classification.
Collapse
Affiliation(s)
- Liam Butler
- Epidemiological Cardiology Research Center, Section on Cardiovascular Medicine, Department of Medicine, Wake Forest School of Medicine, Winston-Salem, North Carolina
| | - Ibrahim Karabayir
- Epidemiological Cardiology Research Center, Section on Cardiovascular Medicine, Department of Medicine, Wake Forest School of Medicine, Winston-Salem, North Carolina
| | - Dalane W. Kitzman
- Epidemiological Cardiology Research Center, Section on Cardiovascular Medicine, Department of Medicine, Wake Forest School of Medicine, Winston-Salem, North Carolina
| | - Alvaro Alonso
- Rollins School of Public Health, Emory University, Atlanta, Georgia
| | - Geoffrey H. Tison
- Division of Cardiology, University of California, San Francisco, California
| | - Lin Yee Chen
- Lillehei Heart Institute and the Department of Medicine (Cardiovascular Division), University of Minnesota Medical School, Minneapolis, Minnesota
| | - Patricia P. Chang
- Department of Medicine (Division of Cardiology), University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
| | - Gari Clifford
- Department of Biomedical Informatics, Emory School of Medicine, Emory University, Atlanta, Georgia
- Wallace H. Coulter Department of Biomedical Engineering, College of Engineering, Georgia Institute of Technology, Atlanta, Georgia
| | - Elsayed Z. Soliman
- Epidemiological Cardiology Research Center, Section on Cardiovascular Medicine, Department of Medicine, Wake Forest School of Medicine, Winston-Salem, North Carolina
| | - Oguz Akbilgic
- Epidemiological Cardiology Research Center, Section on Cardiovascular Medicine, Department of Medicine, Wake Forest School of Medicine, Winston-Salem, North Carolina
| |
Collapse
|
5
|
Wang H, Guo X, Zheng Y, Yang Y. An automatic approach for heart failure typing based on heart sounds and convolutional recurrent neural networks. Phys Eng Sci Med 2022; 45:475-485. [PMID: 35347667 DOI: 10.1007/s13246-022-01112-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/08/2021] [Accepted: 02/18/2022] [Indexed: 11/26/2022]
Abstract
Heart failure (HF) is a complex clinical syndrome that poses a major hazard to human health. Patients with different types of HF have great differences in pathogenesis and treatment options. Therefore, HF typing is of great significance for timely treatment of patients. In this paper, we proposed an automatic approach for HF typing based on heart sounds (HS) and convolutional recurrent neural networks, which provides a new non-invasive and convenient way for HF typing. Firstly, the collected HS signals were preprocessed with adaptive wavelet denoising. Then, the logistic regression based hidden semi-Markov model was utilized to segment HS frames. For the distinction between normal subjects and the HF patients with preserved ejection fraction or reduced ejection fraction, a model based on convolutional neural network and recurrent neural network was built. The model can automatically learn the spatial and temporal characteristics of HS signals. The results show that the proposed model achieved a superior performance with an accuracy of 97.64%. This study suggests the proposed method could be a useful tool for HF recognition and as a supplement for HF typing.
Collapse
Affiliation(s)
- Hui Wang
- Key Laboratory of Biorheology Science and Technology, Ministry of Education, College of Bioengineering, Chongqing University, Chongqing, 400044, China
| | - Xingming Guo
- Key Laboratory of Biorheology Science and Technology, Ministry of Education, College of Bioengineering, Chongqing University, Chongqing, 400044, China.
| | - Yineng Zheng
- Department of Radiology, The First Affiliated Hospital of Chongqing Medical University, Chongqing, 400016, China
| | - Yang Yang
- Key Laboratory of Biorheology Science and Technology, Ministry of Education, College of Bioengineering, Chongqing University, Chongqing, 400044, China
| |
Collapse
|
6
|
Bachtiger P, Petri CF, Scott FE, Ri Park S, Kelshiker MA, Sahemey HK, Dumea B, Alquero R, Padam PS, Hatrick IR, Ali A, Ribeiro M, Cheung WS, Bual N, Rana B, Shun-Shin M, Kramer DB, Fragoyannis A, Keene D, Plymen CM, Peters NS. Point-of-care screening for heart failure with reduced ejection fraction using artificial intelligence during ECG-enabled stethoscope examination in London, UK: a prospective, observational, multicentre study. Lancet Digit Health 2022; 4:e117-e125. [PMID: 34998740 PMCID: PMC8789562 DOI: 10.1016/s2589-7500(21)00256-9] [Citation(s) in RCA: 35] [Impact Index Per Article: 17.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/12/2021] [Revised: 10/21/2021] [Accepted: 11/01/2021] [Indexed: 02/06/2023]
Abstract
Background Most patients who have heart failure with a reduced ejection fraction, when left ventricular ejection fraction (LVEF) is 40% or lower, are diagnosed in hospital. This is despite previous presentations to primary care with symptoms. We aimed to test an artificial intelligence (AI) algorithm applied to a single-lead ECG, recorded during ECG-enabled stethoscope examination, to validate a potential point-of-care screening tool for LVEF of 40% or lower. Methods We conducted an observational, prospective, multicentre study of a convolutional neural network (known as AI-ECG) that was previously validated for the detection of reduced LVEF using 12-lead ECG as input. We used AI-ECG retrained to interpret single-lead ECG input alone. Patients (aged ≥18 years) attending for transthoracic echocardiogram in London (UK) were recruited. All participants had 15 s of supine, single-lead ECG recorded at the four standard anatomical positions for cardiac auscultation, plus one handheld position, using an ECG-enabled stethoscope. Transthoracic echocardiogram-derived percentage LVEF was used as ground truth. The primary outcome was performance of AI-ECG at classifying reduced LVEF (LVEF ≤40%), measured using metrics including the area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity, with two-sided 95% CIs. The primary outcome was reported for each position individually and with an optimal combination of AI-ECG outputs (interval range 0–1) from two positions using a rule-based approach and several classification models. This study is registered with ClinicalTrials.gov, NCT04601415. Findings Between Feb 6 and May 27, 2021, we recruited 1050 patients (mean age 62 years [SD 17·4], 535 [51%] male, 432 [41%] non-White). 945 (90%) had an ejection fraction of at least 40%, and 105 (10%) had an ejection fraction of 40% or lower. Across all positions, ECGs were most frequently of adequate quality for AI-ECG interpretation at the pulmonary position (979 [93·3%] of 1050). Quality was lowest for the aortic position (846 [80·6%]). AI-ECG performed best at the pulmonary valve position (p=0·02), with an AUROC of 0·85 (95% CI 0·81–0·89), sensitivity of 84·8% (76·2–91·3), and specificity of 69·5% (66·4–72·6). Diagnostic odds ratios did not differ by age, sex, or non-White ethnicity. Taking the optimal combination of two positions (pulmonary and handheld positions), the rule-based approach resulted in an AUROC of 0·85 (0·81–0·89), sensitivity of 82·7% (72·7–90·2), and specificity of 79·9% (77·0–82·6). Using AI-ECG outputs from these two positions, a weighted logistic regression with l2 regularisation resulted in an AUROC of 0·91 (0·88–0·95), sensitivity of 91·9% (78·1–98·3), and specificity of 80·2% (75·5–84·3). Interpretation A deep learning system applied to single-lead ECGs acquired during a routine examination with an ECG-enabled stethoscope can detect LVEF of 40% or lower. These findings highlight the potential for inexpensive, non-invasive, workflow-adapted, point-of-care screening, for earlier diagnosis and prognostically beneficial treatment. Funding NHS Accelerated Access Collaborative, NHSX, and the National Institute for Health Research.
Collapse
|
7
|
Tóth N, Soós A, Váradi A, Hegyi P, Tinusz B, Vágvölgyi A, Orosz A, Solymár M, Polyák A, Varró A, Farkas AS, Nagy N. Effect of ivabradine in heart failure: a meta-analysis of heart failure patients with reduced versus preserved ejection fraction. Can J Physiol Pharmacol 2021; 99:1159-1174. [PMID: 34636643 DOI: 10.1139/cjpp-2020-0700] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/22/2022]
Abstract
In clinical trials of heart failure reduced ejection fraction (HFrEF), ivabradine seemed to be an effective heart rate lowering agent associated with lower risk of cardiovascular death. In contrast, ivabradine failed to improve cardiovascular outcomes in heart failure preserved ejection fraction (HFpEF) despite the significant effect on heart rate. This meta-analysis is the first to compare the effects of ivabradine on heart rate and mortality parameters in HFpEF versus HFrEF. We screened three databases: PubMed, Embase, and Cochrane Library. The outcomes of these studies were mortality, reduction in heart rate, and left ventricular function improvement. We compared the efficacy of ivabradine treatment in HFpEF versus HFrEF. Heart rate analysis of pooled data showed decrease in both HFrEF (-17.646 beats/min) and HFpEF (-11.434 beats/min), and a tendency to have stronger bradycardic effect in HFrEF (p = 0.094) in randomized clinical trials. Left ventricular ejection fraction analysis revealed significant improvement in HFrEF (5.936, 95% CI: [4.199-7.672], p < 0.001) when compared with placebo (p < 0.001). We found that ivabradine significantly improves left ventricular performance in HFrEF, at the same time it exerts a tendency to have improved bradycardic effect in HFrEF. These disparate effects of ivabradine and the higher prevalence of non-cardiac comorbidities in HFpEF may explain the observed beneficial effects in HFrEF and the unchanged outcomes in HFpEF patients after ivabradine treatment.
Collapse
Affiliation(s)
- Noémi Tóth
- Department of Pharmacology and Pharmacotherapy, Albert Szent-Györgyi Medical School University of Szeged, Dóm Square 12, Szeged 6720, Hungary
| | - Alexandra Soós
- Institute for Translational Medicine, Medical School, University of Pécs, 12 Szigeti Street, Pécs 7624, Hungary
| | - Alex Váradi
- Institute for Translational Medicine, Medical School, University of Pécs, 12 Szigeti Street, Pécs 7624, Hungary
| | - Péter Hegyi
- Institute for Translational Medicine, Medical School, University of Pécs, 12 Szigeti Street, Pécs 7624, Hungary
| | - Benedek Tinusz
- Institute for Translational Medicine, Medical School, University of Pécs, 12 Szigeti Street, Pécs 7624, Hungary.,First Department of Medicine, Medical School, University of Pécs, Ifjúság Street 13, Pécs 7624, Hungary
| | - Anna Vágvölgyi
- Department of Internal Medicine, Albert Szent-Györgyi Medical School University of Szeged, Kálvária sgt. 57, Szeged 6720, Hungary
| | - Andrea Orosz
- Department of Pharmacology and Pharmacotherapy, Albert Szent-Györgyi Medical School University of Szeged, Dóm Square 12, Szeged 6720, Hungary
| | - Margit Solymár
- Institute for Translational Medicine, Medical School, University of Pécs, 12 Szigeti Street, Pécs 7624, Hungary
| | - Alexandra Polyák
- Department of Internal Medicine, Albert Szent-Györgyi Medical School University of Szeged, Kálvária sgt. 57, Szeged 6720, Hungary
| | - András Varró
- Department of Pharmacology and Pharmacotherapy, Albert Szent-Györgyi Medical School University of Szeged, Dóm Square 12, Szeged 6720, Hungary.,ELKH-SZTE Research Group of Cardiovascular Pharmacology, Szeged, Hungary
| | - Attila S Farkas
- Department of Internal Medicine, Albert Szent-Györgyi Medical School University of Szeged, Kálvária sgt. 57, Szeged 6720, Hungary
| | - Norbert Nagy
- Department of Pharmacology and Pharmacotherapy, Albert Szent-Györgyi Medical School University of Szeged, Dóm Square 12, Szeged 6720, Hungary.,ELKH-SZTE Research Group of Cardiovascular Pharmacology, Szeged, Hungary
| |
Collapse
|
8
|
Vaid A, Johnson KW, Badgeley MA, Somani SS, Bicak M, Landi I, Russak A, Zhao S, Levin MA, Freeman RS, Charney AW, Kukar A, Kim B, Danilov T, Lerakis S, Argulian E, Narula J, Nadkarni GN, Glicksberg BS. Using Deep-Learning Algorithms to Simultaneously Identify Right and Left Ventricular Dysfunction From the Electrocardiogram. JACC Cardiovasc Imaging 2021; 15:395-410. [PMID: 34656465 PMCID: PMC8917975 DOI: 10.1016/j.jcmg.2021.08.004] [Citation(s) in RCA: 27] [Impact Index Per Article: 9.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 04/01/2021] [Revised: 07/29/2021] [Accepted: 08/05/2021] [Indexed: 12/20/2022]
Abstract
OBJECTIVES This study sought to develop DL models capable of comprehensively quantifying left and right ventricular dysfunction from ECG data in a large, diverse population. BACKGROUND Rapid evaluation of left and right ventricular function using deep learning (DL) on electrocardiograms (ECGs) can assist diagnostic workflow. However, DL tools to estimate right- ventricular (RV) function do not exist, whereas those to estimate left ventricular (LV) function are restricted to quantification of very low LV function only. METHODS A multicenter study was conducted with data from 5 New York City hospitals: 4 for internal testing and 1 serving as external validation. We created novel DL models to classify left ventricular ejection fraction (LVEF) into categories derived from the latest universal definition of heart failure, estimate LVEF through regression, and predict a composite outcome of either RV systolic dysfunction or RV dilation. RESULTS We obtained echocardiogram LVEF estimates for 147,636 patients paired to 715,890 ECGs. We used natural language processing (NLP) to extract RV size and systolic function information from 404,502 echocardiogram reports paired to 761,510 ECGs for 148,227 patients. For LVEF classification in internal testing, area under curve (AUC) at detection of LVEF ≤40%, 40% < LVEF ≤50%, and LVEF >50% was 0.94 (95% CI: 0.94-0.94), 0.82 (95% CI: 0.81-0.83), and 0.89 (95% CI: 0.89-0.89), respectively. For external validation, these results were 0.94 (95% CI: 0.94-0.95), 0.73 (95% CI: 0.72-0.74), and 0.87 (95% CI: 0.87-0.88). For regression, the mean absolute error was 5.84% (95% CI: 5.82%-5.85%) for internal testing and 6.14% (95% CI: 6.13%-6.16%) in external validation. For prediction of the composite RV outcome, AUC was 0.84 (95% CI: 0.84-0.84) in both internal testing and external validation. CONCLUSIONS DL on ECG data can be used to create inexpensive screening, diagnostic, and predictive tools for both LV and RV dysfunction. Such tools may bridge the applicability of ECGs and echocardiography and enable prioritization of patients for further interventions for either sided failure progressing to biventricular disease.
Collapse
Affiliation(s)
- Akhil Vaid
- Mount Sinai Clinical Intelligence Center, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, New York, USA; Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Kipp W Johnson
- Mount Sinai Clinical Intelligence Center, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | | | - Sulaiman S Somani
- The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Mesude Bicak
- The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, New York, USA; Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Isotta Landi
- Mount Sinai Clinical Intelligence Center, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, New York, USA; Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, New York, USA; Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Adam Russak
- Department of Anesthesiology, Perioperative and Pain Medicine, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Shan Zhao
- The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, New York, USA; Department of Population Health Science and Policy, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Matthew A Levin
- Mount Sinai Clinical Intelligence Center, Icahn School of Medicine at Mount Sinai, New York, New York, USA; Department of Population Health Science and Policy, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Robert S Freeman
- Mount Sinai Clinical Intelligence Center, Icahn School of Medicine at Mount Sinai, New York, New York, USA; Institute for Healthcare Delivery Science, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Pamela Sklar Division of Psychiatric Genomics, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Alexander W Charney
- Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, New York, USA; Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, New York, USA; Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Atul Kukar
- Department of Cardiology, Mount Sinai Queens Hospital, Astoria, New York, USA, and Icahn School of Medicine at Mount Sinai, New York, New York, USA; Division of Cardiology, Mount Sinai West Hospital and Icahn School of Medicine at Mount Sinai, New York, New York USA
| | - Bette Kim
- Mount Sinai Beth Israel Hospital, Icahn School of Medicine at Mount Sinai, New York, New York, USA; Mount Sinai Heart, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Tatyana Danilov
- Department of Cardiology, Mount Sinai Morningside Hospital, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Stamatios Lerakis
- The Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Edgar Argulian
- Mount Sinai Heart, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Jagat Narula
- The Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Girish N Nadkarni
- Mount Sinai Clinical Intelligence Center, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, New York, USA
| | - Benjamin S Glicksberg
- Mount Sinai Clinical Intelligence Center, Icahn School of Medicine at Mount Sinai, New York, New York, USA; The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, New York, USA; Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, New York, USA.
| |
Collapse
|
9
|
Kwon JM, Kim KH, Eisen HJ, Cho Y, Jeon KH, Lee SY, Park J, Oh BH. Artificial intelligence assessment for early detection of heart failure with preserved ejection fraction based on electrocardiographic features. EUROPEAN HEART JOURNAL. DIGITAL HEALTH 2020; 2:106-116. [PMID: 36711179 PMCID: PMC9707919 DOI: 10.1093/ehjdh/ztaa015] [Citation(s) in RCA: 11] [Impact Index Per Article: 2.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 10/18/2020] [Revised: 11/10/2020] [Accepted: 11/20/2020] [Indexed: 02/01/2023]
Abstract
Aims Although heart failure with preserved ejection fraction (HFpEF) is a rapidly emerging global health problem, an adequate tool to screen it reliably and economically does not exist. We developed an interpretable deep learning model (DLM) using electrocardiography (ECG) and validated its performance. Methods and results This retrospective cohort study included two hospitals. 34 103 patients who underwent echocardiography and ECG within 1 week and indicated normal left ventricular systolic function were included in this study. A DLM based on an ensemble neural network was developed using 32 671 ECGs of 20 169 patients. The internal validation included 1979 ECGs of 1979 patients. Furthermore, we conducted an external validation with 11 955 ECGs of 11 955 patients from another hospital. The endpoint was to detect HFpEF. During the internal and external validation, the area under the receiver operating characteristic curves of a DLM using 12-lead ECG for detecting HFpEF were 0.866 (95% confidence interval 0.850-0.883) and 0.869 (0.860-0.877), respectively. In the 1412 individuals without HFpEF at initial echocardiography, patients whose DLM was defined as having a higher risk had a significantly higher chance of developing HFpEF than those in the low-risk group (33.6% vs. 8.4%, P < 0.001). Sensitivity map showed that the DLM focused on the QRS complex and T-wave. Conclusion The DLM demonstrated high performance for HFpEF detection using not only a 12-lead ECG but also 6- single-lead ECG. These results suggest that HFpEF can be screened using conventional ECG devices and diverse life-type ECG machines employing the DLM, thereby preventing disease progression.
Collapse
Affiliation(s)
- Joon-myoung Kwon
- Department of Critical Care and Emergency Medicine, Mediplex Sejong Hospital, Incheon, South Korea,Artificial Intelligence and Big Data Research Center, Sejong Medical Research Institute, Bucheon, South Korea,Medical Research Team, Medical AI, Co. Seoul, South Korea,Medical R&D Center, Body Friend, Co. Seoul, South Korea
| | - Kyung-Hee Kim
- Artificial Intelligence and Big Data Research Center, Sejong Medical Research Institute, Bucheon, South Korea,Division of Cardiology, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, South Korea,Corresponding author. Tel: 82-32-240-8245, Fax: 82-32-240-8094,
| | - Howard J Eisen
- Penn State Heart and Vascular Institute, Pennsylvania State University/Milton S. Hershey Medical Center, Hershey, PA, USA
| | - Younghoon Cho
- Medical R&D Center, Body Friend, Co. Seoul, South Korea
| | - Ki-Hyun Jeon
- Artificial Intelligence and Big Data Research Center, Sejong Medical Research Institute, Bucheon, South Korea,Division of Cardiology, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, South Korea
| | - Soo Youn Lee
- Artificial Intelligence and Big Data Research Center, Sejong Medical Research Institute, Bucheon, South Korea,Division of Cardiology, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, South Korea
| | - Jinsik Park
- Division of Cardiology, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, South Korea
| | - Byung-Hee Oh
- Division of Cardiology, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, South Korea
| |
Collapse
|
10
|
Wang Y, Ma X. Relationship between changes of electrocardiogram indexes in chronic heart failure with arrhythmia and serum PIIINP and BNP. Exp Ther Med 2020; 19:591-596. [PMID: 31897101 PMCID: PMC6923753 DOI: 10.3892/etm.2019.8269] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/23/2019] [Accepted: 11/20/2019] [Indexed: 01/01/2023] Open
Abstract
Relationship between changes of electrocardiogram (ECG) indexes in chronic heart failure (CHF) with arrhythmia and serum type III procollagen amino-terminal peptide (PIIINP) and brain natriuretic peptide (BNP) were evaluated. From December 2017 to December 2018, 101 patients with heart failure (HF) were collected. Among them, 48 patients with HF and slow arrhythmia were in group A, and 53 cases of HF with non-slow arrhythmia were sin group B, including 33 males and 20 females. BNP was detected by chemiluminescence and PIIINP was detected by immunoassay. The changes of ECG indexes in the two groups, the correlation between serum PIIINP and BNP and NYHA classification of cardiac function, and the correlation between ECG indexes and PIIINP and BNP were detected. ROC curve analysis of BNP and PIIINP in the diagnosis of slow HF was carried out. Serum PIIINP and BNP in group A were significantly higher than those in group B (P<0.05). The levels of PIIINP and BNP in serum of NYHA patients with different cardiac functions, and those in serum of patients with class III were significantly higher than those of group II (P<0.05), while significantly lower than those of group IV (P<0.05). The heart rate and Q-T interval in group A were significantly higher than those in group B (P<0.05). The P-R interval and QES wave group in group A were significantly lower than those in group B (P<0.05). BNP had a positive correlation with Hr and G-T, and was negatively correlated with P-R and QRS; PIIINP was positively correlated with Hr and G-T, and had a negative correlation with P-R, QRS and BNP; PIIINP had positive correlation with NYHA; ECG indexes were correlated with BNP and PIIINP, and had diagnostic value for CHF. Using ECG indexes to predict BNP and PIIINP levels was conducive to the diagnosis of CHF.
Collapse
Affiliation(s)
- Yue Wang
- Electrocardiogram Room, Yantai Yuhuangding Hospital, Yantai, Shandong 264000, P.R. China
| | - Xiaoke Ma
- Departnent of Laboratory Medicine, Yantai Hospital of Traditional Chinese Medicine, Yantai, Shandong 264000, P.R. China
| |
Collapse
|
11
|
Nikolaidou T, Samuel NA, Marincowitz C, Fox DJ, Cleland JGF, Clark AL. Electrocardiographic characteristics in patients with heart failure and normal ejection fraction: A systematic review and meta-analysis. Ann Noninvasive Electrocardiol 2019; 25:e12710. [PMID: 31603593 PMCID: PMC7358891 DOI: 10.1111/anec.12710] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/10/2019] [Revised: 09/03/2019] [Accepted: 09/11/2019] [Indexed: 12/28/2022] Open
Abstract
BACKGROUND Little is known about ECG abnormalities in patients with heart failure and normal ejection fraction (HeFNEF) and how they relate to different etiologies or outcomes. METHODS AND RESULTS We searched the literature for peer-reviewed studies describing ECG abnormalities in HeFNEF other than heart rhythm alone. Thirty five studies were identified and 32,006 participants. ECG abnormalities reported in patients with HeFNEF include atrial fibrillation (prevalence 12%-46%), long PR interval (11%-20%), left ventricular hypertrophy (LVH, 10%-30%), pathological Q waves (11%-18%), RBBB (6%-16%), LBBB (0%-8%), and long JTc (3%-4%). Atrial fibrillation is more common in patients with HeFNEF compared to those with heart failure and reduced ejection fraction (HeFREF). In contrast, long PR interval, LVH, Q waves, LBBB, and long JTc are more common in patients with HeFREF. A pooled effect estimate analysis showed that QRS duration ≥120 ms, although uncommon (13%-19%), is associated with worse outcomes in patients with HeFNEF. CONCLUSIONS There is high variability in the prevalence of ECG abnormalities in patients with HeFNEF. Atrial fibrillation is more common in patients with HeFNEF compared to those with HeFREF. QRS duration ≥120 ms is associated with worse outcomes in patients with HeFNEF. Further studies are needed to address whether ECG abnormalities correlate with different phenotypes in HeFNEF.
Collapse
Affiliation(s)
- Theodora Nikolaidou
- Wythenshawe Hospital, Manchester University NHS Foundation Trust, Manchester, UK
| | - Nathan A Samuel
- Department of Academic Cardiology, Castle Hill Hospital, University of Hull, Hull, UK
| | - Carl Marincowitz
- Hull York Medical School, University of Hull, University of York, York, UK
| | - David J Fox
- Wythenshawe Hospital, Manchester University NHS Foundation Trust, Manchester, UK
| | - John G F Cleland
- Robertson Institute of Biostatistics and Clinical Trials Unit, University of Glasgow, Glasgow, UK.,National Heart & Lung Institute and National Institute of Health Research Cardiovascular Biomedical Research Unit, Imperial College, Royal Brompton & Harefield Hospitals, London, UK
| | - Andrew L Clark
- Department of Academic Cardiology, Castle Hill Hospital, University of Hull, Hull, UK
| |
Collapse
|
12
|
Boriani G, Vitolo M. The 12-lead ECG: a continuous reference for the cardiologist. J Cardiovasc Med (Hagerstown) 2019; 20:459-463. [PMID: 31045692 DOI: 10.2459/jcm.0000000000000803] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/22/2022]
Affiliation(s)
- Giuseppe Boriani
- Cardiology Division, Department of Biomedical, Metabolic and Neural Sciences, University of Modena and Reggio Emilia, Policlinico di Modena, Modena, Italy
| | | |
Collapse
|
13
|
Park J. Can artificial Intelligence Prediction Algorithms Exceed Statistical Predictions? Korean Circ J 2019; 49:640-641. [PMID: 31074231 PMCID: PMC6597450 DOI: 10.4070/kcj.2019.0110] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/04/2019] [Accepted: 04/10/2019] [Indexed: 01/08/2023] Open
Affiliation(s)
- Junbeom Park
- Department of Cardiology, College of Medicine, Ewha Womans University, Seoul, Korea.
| |
Collapse
|
14
|
Mozzini C, Cominacini L, Casadei A, Schiavone C, Soresi M. Ultrasonography in Heart Failure: A Story that Matters. Curr Probl Cardiol 2019; 44:116-136. [PMID: 30172551 DOI: 10.1016/j.cpcardiol.2018.05.003] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/23/2018] [Accepted: 05/11/2018] [Indexed: 02/07/2023]
|
15
|
Kwon JM, Kim KH, Jeon KH, Kim HM, Kim MJ, Lim SM, Song PS, Park J, Choi RK, Oh BH. Development and Validation of Deep-Learning Algorithm for Electrocardiography-Based Heart Failure Identification. Korean Circ J 2019; 49:629-639. [PMID: 31074221 PMCID: PMC6597456 DOI: 10.4070/kcj.2018.0446] [Citation(s) in RCA: 61] [Impact Index Per Article: 12.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/17/2018] [Revised: 01/28/2019] [Accepted: 02/19/2019] [Indexed: 12/11/2022] Open
Abstract
BACKGROUND AND OBJECTIVES Screening and early diagnosis for heart failure (HF) are critical. However, conventional screening diagnostic methods have limitations, and electrocardiography (ECG)-based HF identification may be helpful. This study aimed to develop and validate a deep-learning algorithm for ECG-based HF identification (DEHF). METHODS The study involved 2 hospitals and 55,163 ECGs of 22,765 patients who performed echocardiography within 4 weeks were study subjects. ECGs were divided into derivation and validation data. Demographic and ECG features were used as predictive variables. The primary endpoint was detection of HF with reduced ejection fraction (HFrEF; ejection fraction [EF]≤40%), and the secondary endpoint was HF with mid-range to reduced EF (≤50%). We developed the DEHF using derivation data and the algorithm representing the risk of HF between 0 and 1. We confirmed accuracy and compared logistic regression (LR) and random forest (RF) analyses using validation data. RESULTS The area under the receiver operating characteristic curves (AUROCs) of DEHF for identification of HFrEF were 0.843 (95% confidence interval, 0.840-0.845) and 0.889 (0.887-0.891) for internal and external validation, respectively, and these results significantly outperformed those of LR (0.800 [0.797-0.803], 0.847 [0.844-0.850]) and RF (0.807 [0.804-0.810], 0.853 [0.850-0.855]) analyses. The AUROCs of deep learning for identification of the secondary endpoint was 0.821 (0.819-0.823) and 0.850 (0.848-0.852) for internal and external validation, respectively, and these results significantly outperformed those of LR and RF. CONCLUSIONS The deep-learning algorithm accurately identified HF using ECG features and outperformed other machine-learning methods.
Collapse
Affiliation(s)
- Joon Myoung Kwon
- Department of Emergency Medicine, Mediplex Sejong Hospital, Incheon, Korea
| | - Kyung Hee Kim
- Division of Cardiology, Department of Internal Medicine, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, Korea.
| | - Ki Hyun Jeon
- Division of Cardiology, Department of Internal Medicine, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, Korea
| | - Hyue Mee Kim
- Division of Cardiology, Department of Internal Medicine, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, Korea
| | - Min Jeong Kim
- Division of Cardiology, Department of Internal Medicine, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, Korea
| | - Sung Min Lim
- Division of Cardiology, Department of Internal Medicine, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, Korea
| | - Pil Sang Song
- Division of Cardiology, Department of Internal Medicine, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, Korea
| | - Jinsik Park
- Division of Cardiology, Department of Internal Medicine, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, Korea
| | - Rak Kyeong Choi
- Division of Cardiology, Department of Internal Medicine, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, Korea
| | - Byung Hee Oh
- Division of Cardiology, Department of Internal Medicine, Cardiovascular Center, Mediplex Sejong Hospital, Incheon, Korea
| |
Collapse
|