Abstract
QUANTITATIVE PREDICTION OF MYOCARDIAL RELAXATION FROM SURFACE ELECTROCARDIOGRAM FOR DIAGNOSIS OF LEFT VENTRICULAR DIASTOLIC DYSFUNCTION
Journal of the American College of Cardiology, Vol.75(11 Supplement 1), pp.1070-1070
03/24/2020
DOI: 10.1016/S0735-1097(20)31697-1
Abstract
Background
Myocardial relaxation is recognized to play a major role in the pathophysiology and management of heart failure. We developed machine-learning algorithms that quantitatively estimate myocardial relaxation using body surface signal-processed electrocardiography (spECG).
Methods
This was a prospective multicenter study. Patients were split into the training and test set. Machine-learning models were built using spECG features for predicting the quantitative values of echocardiography-derived tissue Doppler early diastolic relaxation velocity (e’) in the training set and tested in the test set.
Results
Machine learning algorithms showed a significant ability to predict lateral (R2 = 0.66, p <0.001) and septal e’ (R2 = 0.31, p <0.001) in the test cohort. The ECG-estimated lateral e’ predicted LV diastolic dysfunction (AUC 0.825) as well as systolic dysfunction (AUC 0.818). The estimated lateral e’ had incremental values to clinical and traditional ECG findings, with improvements in the AUC and net reclassification, and this combination predicted LV diastolic dysfunction with a high sensitivity of 90.5% for both (AUC 0.856, specificity 66.0% for LV diastolic dysfunction).
Conclusion
A quantitative estimation of myocardial relaxation can be performed using features that are relatively easily obtained using body surface spECG. This cost-effective strategy may be a valuable first clinical step for assessing the presence of LV systolic and diastolic dysfunction.
Details
- Title: Subtitle
- QUANTITATIVE PREDICTION OF MYOCARDIAL RELAXATION FROM SURFACE ELECTROCARDIOGRAM FOR DIAGNOSIS OF LEFT VENTRICULAR DIASTOLIC DYSFUNCTION
- Creators
- Nobuyuki Kagiyama - West Virginia UniversityMarco Piccirilli - West Virginia UniversitySirish Shrestha - West Virginia UniversityPeter Farjo - West Virginia UniversityMarton Tokodi - West Virginia UniversityGrace Casaclang-VerzosaWadea Tarhuni - West Virginia UniversityNegin Nezarat - West Virginia UniversityMatthew J. Budoff - West Virginia UniversityJagat Narula - Icahn School of Medicine at Mount SinaiPartho Sengupta - West Virginia University
- Resource Type
- Abstract
- Publication Details
- Journal of the American College of Cardiology, Vol.75(11 Supplement 1), pp.1070-1070
- DOI
- 10.1016/S0735-1097(20)31697-1
- ISSN
- 0735-1097
- eISSN
- 1558-3597
- Language
- English
- Date published
- 03/24/2020
- Academic Unit
- Internal Medicine
- Record Identifier
- 9984695809302771
Metrics
1 Record Views