HMN 2025: How Real-time AI-driven choice assist aids catheter ablation

An synthetic intelligence (AI)-driven model can considerably enhance procedural security in cardiac electrophysiology with real-time choice assist, based on a research introduced on the annual assembly of the Heart Rhythm Society, held from April 24 to 27 in San Diego.

Chih-Min Liu, M.D., Ph.D., from Taipei Veterans General Hospital in Taiwan, and colleagues developed the DeePRISM model utilizing retrospective information from 110 sufferers with persistent atrial fibrillation (AF), incorporating deep {learning} and PRISM (Morphological Repetitiveness by Periodicity and Similarity) evaluation to investigate intracardiac electrograms and predict AF termination websites throughout catheter ablation. An further 37 sufferers with persistent AF had been used to prospectively validate the model.

The researchers discovered that the DeePRISM model achieved sturdy efficiency within the impartial check set with an space underneath the curve of 0.87 (sensitivity, 74.3%, specificity, 95.4%, total accuracy, 95.4%). Roughly 4 in 10 sufferers (40.5%) achieved acute AF termination when present process DeePRISM-guided ablation. During a imply follow-up of 27.3 months, the DeePRISM group had considerably superior arrhythmia-free survival versus controls (70.3 versus 35.1%). Additionally, DeePRISM-guided ablation considerably diminished the chance for AF recurrence (hazard ratio, 0.10) with none periprocedural problems.

“With the introduction of the DeePRISM model, we’re taking a big step ahead within the therapy of persistent atrial fibrillation,” Liu mentioned in an announcement. “Our study exhibits that AI-driven, real-time evaluation not solely enhances the success of the process but additionally ensures affected person security, marking a promising advance in electrophysiology.”