Key result
Machine learning aims to derive synthetic CMR-LGE scar maps from intraprocedural voltage recordings.
Why the study?
Most patients presenting for VT ablation do not undergo preprocedural CMR, limiting scar characterization and guidance for ablation.
Can machine learning applied to intraprocedural electroanatomic voltage subtraction features generate synthetic scar maps approximating CMR-LGE in patients with ventricular tachycardia?
Can machine learning applied to intraprocedural electroanatomic voltage subtraction features generate synthetic scar maps approximating CMR-LGE in patients with ventricular tachycardia?
Machine learning applied to intraprocedural voltage recordings is hypothesized to generate synthetic scar maps approximating CMR-LGE, potentially overcoming the lack of preprocedural imaging in VT ablation.
May enable intraprocedural scar mapping without preprocedural CMR in VT; hypothesis-generating and requires prospective validation before adoption.
CMR-LGE is the gold standard for scar characterization in VT, and CMR-guided ablation has shown benefit in observational studies. However, most patients presenting for VT ablation do not undergo preprocedural CMR. We hypothesized that machine learning applied to voltage recordings obtained from intraprocedural data alone, specifically unipolar–bipolar subtraction encoding scar depth, could generate synthetic scar maps approximating CMR-LGE.
No takes yet. Share an insight, caveat, or question.
Alnazeer et al. (2026) studied Ventricular tachycardia. Machine learning applied to intraprocedural voltage recordings vs. CMR-LGE was evaluated on Generation of synthetic scar maps approximating CMR-LGE. Machine learning applied to intraprocedural voltage recordings was hypothesized to generate synthetic scar maps approximating CMR-LGE, but no quantitative results were reported in the abstract.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: