Personalized LSTM Models for ECG Lead Transformations Led to Fewer Diagnostic Errors Than Generalized Models: Deriving 12-Lead ECG from Lead II, V2, and V6

Abstract

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.

A Comparison of Personalized and Generalized LSTM Neural Networks for Deriving VCG from 12-Lead ECG

“…We had previously presented the methodology of training a generalized model and then applying transfer learning for a different problem, which was for the S12 lead ECG derivation from a subset of leads, namely Lead II, V2, and V6 [25]. However, in this paper, we evaluate the performance of the transformations from S12 lead to Frank XYZ lead.…”

Visualized Lead Selection for Arrhythmia Classification Based on a Lead Activation Heatmap Using Multi-Lead ECGs

“…Cho et al [ 10 ] used limb 6-lead ECG signals—specifically leads I, II, III, aVR, aVL, and aVF—from 12-lead ECGs to develop a deep learning-based artificial intelligence algorithm (DLA) for detecting myocardial infarction (MI). Kumar et al [ 11 ] selected leads II, V2, and V6 from 12-lead ECG signals to design a LSTM network for cardiovascular disease detection. Li et al [ 12 ] used leads II and V1 from 12-lead ECG signals to develop a 31-layer 1-D residual convolutional neural network for arrhythmia classification.…”