Arrhythmias & electrophysiology
Artificial intelligence–adjudicated spatiotemporal dispersion: A patient-unique fingerprint of persistent atrial fibrillation
Authors' conclusion
The atrial distribution and extent of multipolar electrogram spatiotemporal dispersion follow a nonrandom, albeit patient-unique, distribution in PsAF patients.
Study context
- Population
- Patients with persistent and long-standing persistent AF undergoing ablation with AI-built (VX1 software) spatiotemporal dispersion maps
- Participants
- 78 patients
- Comparison
- Low vs medium vs high AI-adjudicated dispersion extent and distribution
- Primary endpoint
- Patient uniqueness of AI-DED and its independence from common procedural and clinical parameters
Methodological detailsconsecutive patients
- Consecutive patients
Primary result
As reported in the abstract
AI-DED uniqueness and association with clinical descriptors
- AI-DED paired-subregion Euclidean distance between patients (patient-unique), average ± SD
- 5.07 ± 0.60
- Euclidean distance, min
- 2.23
- Euclidean distance, max
- 9.75
- AF type vs AI-DED levels: no correlation
- R2 = 0.006, P = .53
- AF history vs AI-DED levels: no correlation
- R2 = 0.03, P = .25
Further findings
AF termination during the procedure
Long-term freedom from AF/atrial tachycardia after multiple procedures
- AF termination during the procedure: P = .954
- Long-term freedom from AF/atrial tachycardia after multiple procedures: P = .517
Original publication
Artificial intelligence–adjudicated spatiotemporal dispersion: A patient-unique fingerprint of persistent atrial fibrillation. Heart Rhythm. Published 11 January 2024. doi:10.
Summarised from the published abstract; every number is checked against it