Active Brain & Nervous System Heart, Stroke & Blood

OpenEP|NET

In plain English

AI plain-English summary

A single heart procedure generates 15 gigabytes of data—equivalent to more than 20 hours of video—that until now has been nearly impossible to analyse across large groups of patients. Atrial fibrillation affects about one in four people during their lives, yet ablation treatment succeeds only 50–80% of the time, and doctors lack reliable ways to predict who will benefit. The problem is not a lack of data but a lack of tools: electrophysiology data is stored in manufacturer-specific formats, not in central databases, so AF research has fallen behind other fields in using big data and artificial intelligence. The researchers have developed OpenEP, a platform that shrinks each patient’s 15Gb of ablation data to roughly 20Mb—about five songs—making large-scale analysis feasible. This project will establish a nationwide cohort of 2,300 AF patients and develop algorithms to predict three outcomes: whether ablation stops the arrhythmia, how patients report their quality of life and symptoms, and whether they later suffer strokes or other cardiovascular problems. If successful, the work could transform a routine clinical data stream into a diagnostic and prognostic tool, enabling personalised treatment decisions and laying the infrastructure for future randomised trials across the UK and internationally.

View original technical description
Context Atrial fibrillation (AF) is a disorder of the heart rhythm which becomes more common with age, increases risks of stroke and reduces quality of life. It affects ~25% of people during their lives. Although AF has been recognised for over 100 years, treatments for AF are frequently unsuccessful. AF can be treated by a procedure called ablation, in which parts of the heart are cauterised. However, ablation is only successful in preventing AF from coming back in 50-80% of patients. We also don't know how, or even if, ablation influences long term health. During ablation, large amounts of data are collected: up to 15Gb per patient, equivalent to >20 hours of video. This data describes the shape and electrical function of the heart. There is an opportunity to use data routinely collected from a first procedure to help decide what to do later, for example, if AF returns. There is also emerging evidence showing how these data can be used for diagnostic purposes, outwith ablation settings. However, until now there has been no way to analyse these data across large groups of patients. Unlike medical imaging data, electrophysiology data is not stored in central databases and manufacturer-specific file formats have limited analysis at scale. AF research has therefore fallen behind other fields in the use of "big data" and artificial intelligence. Recently, we released the Open Electrophysiology Framework for Research (OpenEP), which reduces electrophysiology/ablation data from 15Gb down to approximately 20Mb - equivalent to about 5 songs of data - per patient. This advance means that the data are now readily accessible for analysis across large groups of thousands of patients with AF. Aims and objectives Here, we seek to establish a nationwide cohort study, with international validation. We will combine our novel platform for extracting markers of atrial health from routine data with a bespoke study app and national data linkage. We will establish a clinical research network, called OpenEP|NET, to address important challenges in the management of AF. In this project, we initially study a sample of 2,300 AF patients to understand who will benefit from AF ablation in three areas: ablation response, patient reported outcomes and cardiovascular outcomes. Ablation response describes whether AF gets better after an ablation procedure. Patient reported outcomes are some of the key things that matter to patients such as quality of life or AF symptoms. Cardiovascular outcomes are the future problems, like stroke, that can lead to long-term consequences. Potential applications and benefits This project will develop algorithms for predicting outcomes and provide the data to support future randomised trials of these technologies. We will engage with patients, funders and the wider public to plan subsequent and wide-reaching clinical studies to test the clinical effectiveness of the algorithms developed here. OpenEP|NET will also begin a new future for electrophysiology research. An important consideration is planning for this future and co-ordination amongst centres. An important part of this project will therefore be focussed on making sure that the right technologies, know-how, infrastructure, access and resources are in place to allow other researchers to make use of OpenEP|NET in the future, both nationally and internationally.

View the original record at the funder ↗

Researchers

Mark O'Neill (Co-Investigator)Michelle Williams (Co-Investigator)Neill Grubb (Co-Investigator)Nicholas Mills (Co-Investigator)Nick Linton (Co-Investigator)Steven Niederer (Co-Investigator)Steven Williams (Principal Investigator)

Related Research

Grants with similar aims, by meaning.

Utilising clinical and electroanatomic mapping data to understand and treat persistent atrial fibrillation
Mapping populations to patients: designing optimal ablation therapy for atrial fibrillation through simulation and deep learning of digital twins
Optimising rate-control therapy in older patients with AF – a training fellowship and pilot study focused on improving quality of life
Towards phenotype-specific diagnostic and treatment strategies in atrial fibrillation and heart failure with preserved ejection fraction
The Role of Learning in Chronic Musculoskeletal Pain

Original classification

Research and Innovation

Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.