ECG-Based Deep Learning and Clinical Risk Factors to Predict Atrial Fibrillation

Khurshid, Shaan; Friedman, Samuel; Reeder, Christopher; Achille, Paolo Di; Diamant, Nathaniel; Singh, Pulkit; Harrington, Lia X.; Wang, Xin; Al-Alusi, Mostafa A; Sarma, Gopal P.; Foulkes, Andrea S.; Ellinor, Patrick T.; Anderson, Christopher; Ho, Jennifer E.; Philippakis, Anthony; Batra, Puneet; Lubitz, Steven A.

doi:10.1161/circulationaha.121.057480

Cited by 146 publications

(106 citation statements)

References 40 publications

(78 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We previously observed that ECG-AI estimates for AF risk are largely influenced by the P wave, a period corresponding to atrial depolarization and repolarization. 7 Moreover, we have previously reported that both ECG-AI and clinical risk for AF are complementary. 7 Here, we extend these observations by identifying genetic signals that have been associated with P wave duration, 36 and documenting the distinct genetic profiles underlying risk estimates generated by ECG-AI and a clinical risk factor model.…”

Section: Discussionmentioning

confidence: 99%

“…The development and validation of the ECG-AI model has previously been reported. 7 Briefly, we trained a convolutional neural network (CNN) to predict 5-year risk of AF using 12-lead ECGs in a longitudinal patient cohort derived from the MGB network. 11 ECG-AI uses an encoding and loss function that takes into account both time to event (i.e., AF) and missingness introduced by censoring (death or loss of follow-up) to estimate a 5-year survival probability of AF.…”

Section: Predicted Risk Of Afmentioning

confidence: 99%

“…The P wave was selected because ( 1) it has been linked to SCN5A/SCN10A in previous genetic studies, 26 and (2) it had the greatest impact on ECG-AI predicted risk indicated by saliency maps and a median waveform analysis in our previous study. 7 Body height was selected because ( 1) it has been linked to VGLL2 and EXT1 loci, 32,33 and ( 2) is an established risk factor for AF. 34,35 A significant and plausible causal effect of P wave duration on ECG-AI risk was supported by four out of the five methods we used, with MR effect sizes ranging from 0.017 to 0.023.…”

Section: Contributing Components Of Ecg-ai Predicted Af Riskmentioning

confidence: 99%

“…8,13,14 The ECG-AI and CHARGE-AF score linear predictors have previously been reported to have modest correlation (Pearson r 0.66) in an independent dataset, and the CHARGE-AF score exhibits similar predictive performance as ECG-AI. 7 For survival analysis in the present study, incident AF at 5 years was ascertained using follow-up information and a previously published definition comprising inpatient ICD-9 and -10 codes, and procedural codes. 15 Follow-up information in the UK Biobank depends on the availability of linked electronic health records and hence varies by enrollment sites.…”

Section: Predicted Risk Of Afmentioning

confidence: 99%

See 3 more Smart Citations

Genetic Susceptibility to Atrial Fibrillation Identified via Deep Learning of 12-lead Electrocardiograms

Wang

Khurshid

Choi

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Artificial intelligence (AI) models applied to 12-lead electrocardiogram (ECG) waveforms can predict atrial fibrillation (AF), a heritable and morbid arrhythmia. We hypothesized that there may be a genetic basis for ECG-AI based risk estimates. We applied an ECG-AI model for predicting incident AF to ECGs from 39,986 UK Biobank participants without AF. We then performed a genome-wide association study (GWAS) of the predicted AF risk. We identified three signals (P<5E-8) at established AF susceptibility loci marked by the sarcomeric gene TTN, and sodium channel genes SCN5A and SCN10A. We also identified two novel loci near the genes VGLL2 and EXT1. In contrast, a GWAS of risk estimates from a clinical variable model indicated a different genetic profile. Predicted AF risk from an ECG-AI model is influenced by genetic variation implicating sarcomeric, ion channel, and height pathways. ECG-AI models may identify individuals at risk for disease via specific biological pathways.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Predicted Risk Of Afmentioning

confidence: 99%

Section: Contributing Components Of Ecg-ai Predicted Af Riskmentioning

confidence: 99%

Section: Predicted Risk Of Afmentioning

confidence: 99%

See 2 more Smart Citations

Genetic Susceptibility to Atrial Fibrillation Identified via Deep Learning of 12-lead Electrocardiograms

Wang

Khurshid

Choi

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Along with the continuous innovation of convolutional algorithms, from LeNet by Lecun et al ( 36 ) to ResNet by He et al ( 37 ), computer-aided decision making through imaging has become a hot topic in medical research, such as prediction of BMI by facial image features to predict BMI ( 38 ) and fundus images to predict diabetic retinopathy ( 39 ). In addition, the prevalence of electronic medical records and the establishment of large medical databases have also provided the basis for research on clinical problems, and the combination with machine learning has shown remarkable performance in predicting the occurrence and prognosis of diseases ( 40 , 41 ). Unfortunately, however, there is still a lack of research in the current field for our patient population.…”

Section: Discussionmentioning

confidence: 99%

Development and Internal Validation of a Nomogram to Predict Mortality During the ICU Stay of Thoracic Fracture Patients Without Neurological Compromise: An Analysis of the MIMIC-III Clinical Database

Wang

Fan

et al. 2021

Front. Public Health

View full text Add to dashboard Cite

Background: This study aimed to develop and validate a nomogram for predicting mortality in patients with thoracic fractures without neurological compromise and hospitalized in the intensive care unit.Methods: A total of 298 patients from the Medical Information Mart for Intensive Care III (MIMIC-III) database were included in the study, and 35 clinical indicators were collected within 24 h of patient admission. Risk factors were identified using the least absolute shrinkage and selection operator (LASSO) regression. A multivariate logistic regression model was established, and a nomogram was constructed. Internal validation was performed by the 1,000 bootstrap samples; a receiver operating curve (ROC) was plotted, and the area under the curve (AUC), sensitivity, and specificity were calculated. In addition, the calibration of our model was evaluated by the calibration curve and Hosmer-Lemeshow goodness-of-fit test (HL test). A decision curve analysis (DCA) was performed, and the nomogram was compared with scoring systems commonly used during clinical practice to assess the net clinical benefit.Results: Indicators included in the nomogram were age, OASIS score, SAPS II score, respiratory rate, partial thromboplastin time (PTT), cardiac arrhythmias, and fluid-electrolyte disorders. The results showed that our model yielded satisfied diagnostic performance with an AUC value of 0.902 and 0.883 using the training set and on internal validation. The calibration curve and the Hosmer-Lemeshow goodness-of-fit (HL). The HL tests exhibited satisfactory concordance between predicted and actual outcomes (P = 0.648). The DCA showed a superior net clinical benefit of our model over previously reported scoring systems.Conclusion: In summary, we explored the incidence of mortality during the ICU stay of thoracic fracture patients without neurological compromise and developed a prediction model that facilitates clinical decision making. However, external validation will be needed in the future.

show abstract

Deep Learning of Electrocardiograms in Sinus Rhythm From US Veterans to Predict Atrial Fibrillation

Yuan,

Duffy,

Dhruva

et al. 2023

JAMA Cardiol

View full text Add to dashboard Cite

ImportanceEarly detection of atrial fibrillation (AF) may help prevent adverse cardiovascular events such as stroke. Deep learning applied to electrocardiograms (ECGs) has been successfully used for early identification of several cardiovascular diseases.ObjectiveTo determine whether deep learning models applied to outpatient ECGs in sinus rhythm can predict AF in a large and diverse patient population.Design, Setting, and ParticipantsThis prognostic study was performed on ECGs acquired from January 1, 1987, to December 31, 2022, at 6 US Veterans Affairs (VA) hospital networks and 1 large non-VA academic medical center. Participants included all outpatients with 12-lead ECGs in sinus rhythm.Main Outcomes and MeasuresA convolutional neural network using 12-lead ECGs from 2 US VA hospital networks was trained to predict the presence of AF within 31 days of sinus rhythm ECGs. The model was tested on ECGs held out from training at the 2 VA networks as well as 4 additional VA networks and 1 large non-VA academic medical center.ResultsA total of 907 858 ECGs from patients across 6 VA sites were included in the analysis. These patients had a mean (SD) age of 62.4 (13.5) years, 6.4% were female, and 93.6% were male, with a mean (SD) CHA2DS2-VASc (congestive heart failure, hypertension, age, diabetes mellitus, prior stroke or transient ischemic attack or thromboembolism, vascular disease, age, sex category) score of 1.9 (1.6). A total of 0.2% were American Indian or Alaska Native, 2.7% were Asian, 10.7% were Black, 4.6% were Latinx, 0.7% were Native Hawaiian or Other Pacific Islander, 62.4% were White, 0.4% were of other race or ethnicity (which is not broken down into subcategories in the VA data set), and 18.4% were of unknown race or ethnicity. At the non-VA academic medical center (72 483 ECGs), the mean (SD) age was 59.5 (15.4) years and 52.5% were female, with a mean (SD) CHA2DS2-VASc score of 1.6 (1.4). A total of 0.1% were American Indian or Alaska Native, 7.9% were Asian, 9.4% were Black, 2.9% were Latinx, 0.03% were Native Hawaiian or Other Pacific Islander, 74.8% were White, 0.1% were of other race or ethnicity, and 4.7% were of unknown race or ethnicity. A deep learning model predicted the presence of AF within 31 days of a sinus rhythm ECG on held-out test ECGs at VA sites with an area under the receiver operating characteristic curve (AUROC) of 0.86 (95% CI, 0.85-0.86), accuracy of 0.78 (95% CI, 0.77-0.78), and F1 score of 0.30 (95% CI, 0.30-0.31). At the non-VA site, AUROC was 0.93 (95% CI, 0.93-0.94); accuracy, 0.87 (95% CI, 0.86-0.88); and F1 score, 0.46 (95% CI, 0.44-0.48). The model was well calibrated, with a Brier score of 0.02 across all sites. Among individuals deemed high risk by deep learning, the number needed to screen to detect a positive case of AF was 2.47 individuals for a testing sensitivity of 25% and 11.48 for 75%. Model performance was similar in patients who were Black, female, or younger than 65 years or who had CHA2DS2-VASc scores of 2 or greater.Conclusions and RelevanceDeep learning of outpatient sinus rhythm ECGs predicted AF within 31 days in populations with diverse demographics and comorbidities. Similar models could be used in future AF screening efforts to reduce adverse complications associated with this disease.

show abstract

ECG-Based Deep Learning and Clinical Risk Factors to Predict Atrial Fibrillation

Cited by 146 publications

References 40 publications

Genetic Susceptibility to Atrial Fibrillation Identified via Deep Learning of 12-lead Electrocardiograms

Genetic Susceptibility to Atrial Fibrillation Identified via Deep Learning of 12-lead Electrocardiograms

Development and Internal Validation of a Nomogram to Predict Mortality During the ICU Stay of Thoracic Fracture Patients Without Neurological Compromise: An Analysis of the MIMIC-III Clinical Database

Deep Learning of Electrocardiograms in Sinus Rhythm From US Veterans to Predict Atrial Fibrillation

Contact Info

Product

Resources

About