Key result
AI-ECG interpretation shows good diagnostic performance for detecting Brugada syndrome with an AUC of 0.88.
Why the study?
AI-ECG interpretation has emerged as a promising approach to identify Brugada syndrome, but its diagnostic viability and clinical applicability required investigation.
Does artificial intelligence electrocardiogram (AI-ECG) interpretation accurately detect Brugada syndrome?
Meta-Analysis
Does artificial intelligence electrocardiogram (AI-ECG) interpretation accurately detect Brugada syndrome?
Effect estimate: AUC 0.88 (95% CI 0.85-0.91)
AI-ECG interpretation demonstrates good diagnostic performance for detecting Brugada syndrome, particularly as a rule-in tool for clinically suspected or high-risk populations.
AI-ECG shows moderate accuracy for BrS detection; extends pooled diagnostic data but leaves open need for prospective validation before clinical adoption.
INTRODUCTION: Artificial intelligence electrocardiogram (AI-ECG) interpretation has emerged as a promising approach to identify Brugada syndrome (BrS). This review sought to investigate diagnostic viability and clinical applicability of AI-ECG interpretation models for detecting BrS. METHODS: A systematic search (PubMed, Scopus, and ScienceDirect) was conducted in November 2025. STATA/BE (v17.0) was used to pool the overall metrics of the quantitative diagnostic test accuracy meta-analysis. RESULTS: Seven studies comprising 15 AI-model analyses were included. AI-ECG interpretation yielded a pooled sensitivity of 0.82 [0.77-0.87], specificity of 0.81 [0.73-0.87], negative likelihood ratio of 0.22 [0.16-0.29], positive likelihood ratio of 4.31 [2.94-6.33], and area under the curve of 0.88 [0.85-0.91]. Clinical applicability assessed using Fagan's nomogram showed minimal diagnostic value for general population screening (pretest probability 0.05%). However, in case-finding (5%) and high-risk scenarios (20%), positive results substantially increased posttest probability (to 18% and 52%, respectively), while negative results reduced it to 1% and 5%. Overall, the Fagan plot indicated a beneficial trend toward rule-in utility rather than rule-out utility. CONCLUSION: AI-ECG interpretation demonstrated a good overall diagnostic performance for detecting BrS and may serve as valuable decision-support assisting clinicians in clinically suspected or high-risk populations. PROTOCOL REGISTRATION: www.crd.york.ac.uk/prospero identifier is CRD420261298566.
No takes yet. Share an insight, caveat, or question.
Cuandra et al. (2026) conducted a meta-analysis in Brugada syndrome. Artificial intelligence electrocardiogram (AI-ECG) interpretation was evaluated on Detection of Brugada syndrome (AUC 0.88, 95% CI 0.85-0.91). AI-ECG interpretation demonstrated good diagnostic performance for detecting Brugada syndrome, with a pooled sensitivity of 0.82, specificity of 0.81, and AUC of 0.88 (95% CI 0.85-0.91).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: