Key result
Personalized LSTM models cut 3-to-12-lead ECG derivation error ~65% versus generalized models.
Why the study?
Multi-lead wearable ECGs are desirable for identifying cardiovascular diseases early, motivating methods to derive multiple leads from a reduced subset of electrodes and comparing personalized to generalized derivation models.
Do personalized LSTM models reduce diagnostic errors compared to generalized models when deriving 12-lead ECGs from Lead II, V2, and V6?
Do personalized LSTM models reduce diagnostic errors compared to generalized models when deriving 12-lead ECGs from Lead II, V2, and V6?
Personalized LSTM models for deriving 12-lead ECGs from a 3-lead subset significantly reduce diagnostic errors and improve signal accuracy compared to generalized models.
Personalized ECG synthesis may aid select cases; leaves open prospective validation before any practice change.
Background and Objective: The prevalence of chronic cardiovascular diseases (CVDs) has risen globally, nearly doubling from 1990 to 2019. ECG is a simple, non-invasive measurement that can help identify CVDs at an early and treatable stage. A multi-lead ECG, up to 15 leads in a wearable form factor, is desirable. We seek to derive multiple ECG leads from a select subset of leads so that the number of electrodes can be reduced in line with a patient-friendly wearable device. We further compare personalized derivations to generalized derivations. Methods: Long-Short Term Memory (LSTM) networks using Lead II, V2, and V6 as input are trained to obtain generalized models using Bayesian Optimization for hyperparameter tuning for all patients and personalized models for each patient by applying transfer learning to the generalized models. We compare quantitatively using error metrics Root Mean Square Error (RMSE), R2, and Pearson correlation (ρ). We compare qualitatively by matching ECG interpretations of board-certified cardiologists. Results: ECG interpretations from personalized models, when corrected for an intra-observer variance, were identical to the original ECGs, whereas generalized models led to errors. Mean performance values for generalized and personalized models were (RMSE-74.31 µV, R2-72.05, ρ-0.88) and (RMSE-26.27 µV, R2-96.38, ρ-0.98), respectively. Conclusions: Diagnostic accuracy based on derived ECG is the most critical validation of ECG derivation methods. Personalized transformation should be sought to derive ECGs. Performing a personalized calibration step to wearable ECG systems and LSTM networks could yield ambulatory 15-lead ECGs with accuracy comparable to clinical ECGs.
No takes yet. Share an insight, caveat, or question.
Kumar et al. (2023) studied Cardiovascular diseases. Personalized LSTM models vs. Generalized LSTM models was evaluated on Quantitative error metrics (RMSE, R2, Pearson correlation) and qualitative ECG interpretations. Personalized LSTM models for deriving 12-lead ECGs from 3 leads yielded superior accuracy (RMSE 26.27 µV vs 74.31 µV) compared to generalized models, with identical clinical interpretations.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: