Localization of ventricular activation origin using patient-specific geometry: Preliminary results

Area of Science:

• Cardiology
• Medical Imaging
• Computational Biology

Background:

• Catheter ablation is a key treatment for ventricular tachycardia (VT).
• Accurate localization of the VT exit site is crucial for successful ablation.
• Current methods may be complex or require extensive pre-procedural data.

Purpose of the Study:

• To evaluate a novel statistical pace-mapping method for VT exit site localization.
• To compare its performance against a traditional electrocardiographic inverse solution.

Main Methods:

• Utilized intraprocedural body-surface potential mapping (BSPM) in 7 VT patients undergoing electroanatomic mapping.
• Tested a multiple linear regression (MLR) based statistical method using 12-lead ECG and patient-specific geometry.
• Compared MLR with a classical deterministic inverse solution requiring BSPM and CT-derived geometry.

Main Results:

• The MLR method demonstrated significantly lower localization error compared to the inverse solution for both epicardial (10.6 vs. 27.3 mm) and endocardial (7.7 vs. 17.1 mm) VT exit sites.
• Pooled analysis confirmed MLR's superior accuracy (P = 0.005).
• MLR required 10-15 pacing sites with known origins to derive reliable patient-specific regression equations.

Conclusions:

• The novel statistical pace-mapping approach provides accurate VT origin localization.
• It is easily implementable and suitable for real-time clinical applications.
• Offers a valuable supplement or alternative to pre-procedure inverse solutions.