Publication

Prediction of Left Atrial Volume Parameters from Resting ECGs and Tabular Data Using Deep Learning in the UK Biobank

2026

Abstract

We present a deep learning model that predicts left atrial (LA) volume from standard 12-lead ECG recordings and basic patient data. This approach offers a low-cost, scalable alternative to MRI-based LA volume measurement, which remains the clinical gold standard but is often inaccessible. Our model performs regression directly on LA volume targets and leverages Shapley values to provide interpretable feature importance. Results highlight the predictive value of ECG signals and demonstrate that patient features such as weight and height contribute meaningfully to the estimation.

Official source

Graph Chat

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

2026

Abstract

Official source

Related concepts (32)

Electrocardiography

Electrocardiography is the process of producing an electrocardiogram (ECG or EKG ), a recording of the heart's electrical activity through repeated cardiac cycles. It is an electrogram of the heart which is a graph of voltage versus time of the electrical activity of the heart using electrodes placed on the skin. These electrodes detect the small electrical changes that are a consequence of cardiac muscle depolarization followed by repolarization during each cardiac cycle (heartbeat).

Signal-averaged electrocardiogram

Signal-averaged electrocardiography (SAECG) is a special electrocardiographic technique, in which multiple electric signals from the heart are averaged to remove interference and reveal small variations in the QRS complex, usually the so-called "late potentials". These may represent a predisposition towards potentially dangerous ventricular tachyarrhythmias. A resting electrocardiogram (ECG) is recorded in the supine position using an ECG machine equipped with SAECG software; this can be done by a physician, nurse, or medical technician.

Cardiac conduction system

The cardiac conduction system (CCS) (also called the electrical conduction system of the heart) transmits the signals generated by the sinoatrial node – the heart's pacemaker, to cause the heart muscle to contract, and pump blood through the body's circulatory system. The pacemaking signal travels through the right atrium to the atrioventricular node, along the bundle of His, and through the bundle branches to Purkinje fibers in the walls of the ventricles. The Purkinje fibers transmit the signals more rapidly to stimulate contraction of the ventricles.

Related publications (20)

Prediction of Left Atrial Volume Parameters from Resting ECGs and Tabular Data Using Deep Learning in the UK Biobank

Chat with Graph Search

Estimation of Physiological Metrics from Resting ECGs Using Deep Learning in the UK Biobank, Including submaximal exercise derived V̇O 2 max, Body Fat Percentage, and Grip Strength

Melanoma Detection through Combining Reinforcement Learning, Generative Adversarial Network, and Bayesian Optimization

Optimizing hepatitis C diagnosis through reinforcement learning feature selection and multi-model machine learning evaluation