AI Hearts: Atrial Fibrillation Treatment Breakthrough
- An artificial intelligence (AI) tool can now forecast the success of heart procedures by using synthetic data, according to a new study. This innovation reduces the reliance on...
- Researchers at Queen Mary University of London developed the AI, which creates medically accurate, synthetic models of fibrotic heart tissue.Fibrosis, or heart scarring, can result from aging, stress,...
- Scar tissue disrupts the heart's electrical system, causing AF.
AI Predicts heart Procedure Success Using Synthetic Data
An artificial intelligence (AI) tool can now forecast the success of heart procedures by using synthetic data, according to a new study. This innovation reduces the reliance on real patient information.
Researchers at Queen Mary University of London developed the AI, which creates medically accurate, synthetic models of fibrotic heart tissue.Fibrosis, or heart scarring, can result from aging, stress, or atrial fibrillation (AF). The study, published in Frontiers in Cardiovascular Medicine, suggests this could personalize care for AF patients, who suffer from irregular heartbeats.
Scar tissue disrupts the heart’s electrical system, causing AF. Doctors currently use MRI scans to assess the scarring pattern, which influences treatment outcomes. Ablation,a common AF treatment,involves creating controlled scars to block erratic signals. Though, success rates vary, making it difficult to predict the best approach for each patient. Access to patient imaging data has also been limited, hindering AI development in this area.
Dr. alexander Zolotarev of Queen Mary University of London, the study’s first author, said, “LGE-MRI provides vital information about heart fibrosis, but obtaining enough scans for comprehensive AI training is challenging.”
The AI model was trained using 100 real LGE-MRI scans from AF patients. It then generated 100 synthetic fibrosis patterns mimicking real heart scarring. These virtual models simulated different ablation strategies across various patient anatomies. The team’s diffusion model produced synthetic fibrosis distributions that closely matched real patient data. Predictions using AI-created patterns proved nearly as reliable as those using genuine patient data. This method protects patient privacy while enabling study of a broader range of cardiac scenarios.
The research emphasizes AI’s role as a clinical support tool. “This isn’t about replacing doctors’ judgement,” Zolotarev said.”It’s about providing clinicians with a refined simulator — allowing them to test different treatment approaches on a digital model of each patient’s unique heart structure before performing the actual procedure.”
Dr. Caroline Roney of queen Mary University of London, lead author of the study, said, “We’re very excited about this research as it addresses the challenge of limited clinical data for cardiac digital twin models…aimed at creating more personalised treatments for atrial fibrillation patients.” This work is part of Roney’s project to develop personalized ‘digital twin’ heart models for AF patients.
With atrial fibrillation affecting 1.4 million people in the U.K. and ablation failing in half of cases,this technology could significantly reduce repeat procedures.The AI approach addresses limited patient data availability and the need to protect sensitive medical information.
What’s next
Future research will focus on refining the AI model and expanding its capabilities to simulate a wider range of cardiac conditions and treatment options. The ultimate goal is to integrate this technology into clinical practice,providing doctors with a powerful tool to personalize treatment plans and improve outcomes for patients with atrial fibrillation.