Neural Network-Based Signal Translation with Application to the ECG
Mohamed Amine Abdelmadjid, Mounir Boukadoum · 2022 20th IEEE Interregional NEWCAS Conference (NEWCAS) · 2022
We describe a machine learning approach to translate physiological signals from one representation to another, with application to the electrocardiogram (ECG). A neural network architecture is presented to transform the raw ECG measurements provided by a novel 11-channel contactless capacitive sensor mat into the standard 12-channel wet electrode ECG. To do this, an adapted version of the CycleGAN model, a bidirectional derivative of the generative adversarial network (GAN) architecture, is developed and evaluated with 31 parallel recordings of cardiac activity using the capacitive mat and a standard wet electrode set up. The mean quadratic error between the standard 12-lead ECG derivations and the corresponding translated mat signals by the proposed CycleGAN model and three other shows neural network architectures used for comparison the method’s effectiveness.