Digital twin technology shows ability to personalize ablation planning
Digital twin technology could reshape electrophysiology (EP) by allowing clinicians to test and optimize ablation strategies on a virtual replica of an individual patient’s heart before treating them. The concept of creating digital twins of patients has evolved from a vague concept to something that is actually being used for clinical research.
The results so far are very promising, explained Natalia Trayanova, PhD, professor of biomedical engineering and director of AI research in health and medicine at the Data Science and AI Institute at Johns Hopkins University. She discussed her team's research and data from other recent studies with Cardiovascular Business in a new video interview.
Digital twins are distinct from conventional artificial intelligence (AI), Trayanova emphasized. While machine-learning algorithms identify patterns across large populations, a digital twin represents the behavior of a specific patient’s organ and is intended to support personalized treatment.
Her team builds digital hearts from patient-specific data to evaluate arrhythmia risk and determine optimal ablation strategies. Applications have included persistent atrial fibrillation (AFib), particularly in patients with substantial substrate remodeling and fibrosis, as well as ventricular tachycardia (VT) in patients with ischemic cardiomyopathy and other forms of cardiac disease.
The approach has moved into early clinical testing. Trayanova highlighted a recently published FDA-approved study involving 10 patients in which MRI scans were used to construct a digital twin before an ablation procedure. The model incorporated scar, fibrosis and border-zone tissue and simulated electrical behavior from the cellular level through the entire heart to identify potential ablation targets.
A key feature of the technology is that the researchers do not simply identify the arrhythmogenic substrate visible at the time of treatment; they virtually ablate the digital twin. This enables testing of the best ablation strategy and then enables a reassessment of the heart to determine whether new potential targets could emerge after the initial lesions are created.
“This is done with the intention to decrease any redo ablations,” Trayanova explained. The goal is to anticipate how the electrical landscape could evolve after treatment and incorporate those future targets into the final ablation plan.
The small study reported no arrhythmia recurrence at the end of one year of follow-up. Eight patients were off antiarrhythmic medications, while two were taking reduced doses. Trayanova described the results as an early demonstration of a highly novel approach while acknowledging that the study was small and conducted at a single center.
The distinction from other approaches to mapping is that the digital twin is designed to account for what happens after ablation rather than simply determining where abnormal electrical activity exists at the present time. The strategy could potentially identify areas that become arrhythmogenic months or even a year after treatment, with the objective of reducing repeat hospitalizations and procedures.
Trayanova said the same concept is being investigated in AFib, where the objective is not simply to ablate AFib, but to identify and treat the substrate capable of sustaining reentrant activity. Her team has completed the OPTIMA randomized clinical trial, an FDA-approved single-center study that enrolled 67 patients comparing pulmonary vein isolation with the digital-twin-guided approach. Two-year follow-up is underway, but early results have shown separation between the treatment groups and virtually no post-ablation tachycardias with the digital-twin strategy, she said.
The broader concept is to treat the virtual heart first, validate the resulting lesion pattern through simulation, and only then export the targets to the electro-anatomical mapping system for use during the actual procedure.
Trayanova said the technology is being studied across a range of arrhythmias and cardiomyopathies, including nonischemic cardiomyopathy and arrhythmogenic right ventricular cardiomyopathy. The approach also could produce smaller, more precisely targeted lesions because the targets are selected through patient-specific mechanistic simulations.
Dave Fornell has covered healthcare for more than 17 years, with a focus in cardiology and radiology. Fornell is a 5-time winner of a Jesse H. Neal Award, the most prestigious editorial honors in the field of specialized journalism. The wins included best technical content, best use of social media and best COVID-19 coverage. Fornell was also a three-time Neal finalist for best range of work by a single author. He produces more than 100 editorial videos each year, most of them interviews with key opinion leaders in medicine. He also writes technical articles, covers key trends, conducts video hospital site visits, and is very involved with social media. E-mail: [email protected]