Skip to main content

The heart’s twin – how pioneering digital twin research is personalising the treatment of Atrial Fibrillation

Current treatments for atrial fibrillation, including ablation, have shown mixed results. Each patient responds differently and it is difficult to predict the best possible course of long-term treatment. Now Dr Caroline Roney’s pioneering research using digital twins – a virtual representation of a real-life object – has earned her a UKRI Future Leaders Fellowship.

Atrial fibrillation (AFib) is the most common form of heart arrhythmia, affecting around 1.4 million people in the UK. If it is not successfully treated, the condition can lead to stroke or heart failure. 

While AFib can sometimes be treated with a combination of medication and lifestyle modification, many patients need ablation, a technique where heat or cold energy is used on diseased areas of the heart. This creates tiny scars which block irregular electrical signals and restore a more typical heartbeat. 

Because management is currently predicated on what works for the average patient, there is a high incidence of recurrence: around 50 percent of patients will have further episodes after initial treatment. 

Many current plans are focused on the short-term reversal of the condition, and do not take into account the likelihood of recurrence, or the long-term outcomes for the patient. Dr Roney is interested in being able to predict how a patient might respond over a longer period of time, and thus how to develop treatment that would prevent AFib happening again. 

Using digital twin technology, researchers are able to more accurately predict a patient’s response to treatment, and plan bespoke intervention where necessary. Dr Roney’s work combines machine learning with digital twins to develop new therapies.

What is a digital twin?

A digital twin is a virtual representation of an object or a system. Using imaging, it represents its real-life twin exactly, and is updated with real-time data over its life cycle. Researchers can use simulation, machine learning and reasoning to help improve the accuracy of the ‘twin. In healthcare, digital twin technology will allow doctors to better predict our unique future health problems and will help them to develop bespoke, effective treatment and lifestyle plans.

The digital twin allows the researchers to test and develop treatment approaches that are specific to that patient. Most importantly, this is risk-free for the patient themselves. Doctors are able to test huge numbers of treatments, which gives a better ‘real-world’ outcome for the patient.

In Dr Roney’s case, she has set out to build a patient-specific model of a person’s heart, which is created through imaging technology. This is then personalised using the patient’s electrical data. Dr Roney’s research also incorporates data across population groups, which helps researchers to better predict how the patient may respond to treatment over a longer time scale.

How was the initial research conducted?

Dr Roney’s team worked with 100 atrial fibrillation patients with a variety of diagnoses who were undergoing their first ablation:

  • 43 paroxysmal patients (a form of AFib which stops on its own or with intervention within seven days)
  • 41 persistent patients (AFib lasts longer than a week, but either stops spontaneously or needs treatment)
  • 16 long-standing persistent (a form of AFib that has lasted for more than a year and doesn’t go away)

The patients were followed for a year after ablation, and their ECG was monitored as they moved around in their day-to-day lives. Along with imaging, this allowed the researchers to build patient-specific biophysical models which combined differing fibrosis patterns, fiber orientation maps, electrical properties, and ablation patterns.

In this way, they could observe the unique atrial properties of each patient, and then use this information to test the ability of the tissue to sustain fibrillation.

The machine learning classifiers used features from the patient history, imaging, and atrial fibrillation simulation metrics to predict when, where and how AFib might recur.

Adding patient-specific data to the mix allowed the researchers to improve the long-term predictions for these patients by 30%.

Virtual twins will benefit everyone from patients to doctors and industry. Better treatment, better-run hospitals, and support for healthcare workers are just a few of the many rewards.
— Dr Caroline Roney

What’s next for digital twin research?

The team’s excellent results led to the award of Dr Roney’s UKRI Future Leaders Fellowship, collaborating with hospitals, universities and the pharmaceutical industry across the world.

It has also led to Queen Mary’s participation in Ecosystem for Digital Twins in Healthcare (EDITH), which aims to accelerate digital twins in healthcare in Europe.

The Queen Mary team is interdisciplinary, working across schools and institutes. This includes the School of Engineering and Materials Science (Dr Roney), Electronic Engineering and Computer Science and the Digital Environment Research Institute (Professor Greg Slabaugh), the William Harvey Research Institute (Professor Steffen Petersen and Dr Aaron Lee), and the Cardiovascular Devices Hub (Professor Anthony Mathur).

Digital twin research in medicine has lagged behind advances in the field in engineering and other disciplines, but Dr Roney’s work offers hope for AFib patients and heralds a new dawn in personalised healthcare.

Valuable collaborations

Dr Roney is grateful for the valuable collaborations she has had with:

As well as other partners within Queen Mary, Barts and The London School of Medicine and Dentistry and beyond:

Decorative image of electrical wires

Digital, information, data

Shaping a healthier, more prosperous and equitable future for our digital society.
Back to top