Automatic wide complex tachycardia differentiation using mathematically synthesized vectorcardiogram signals.

Area of Science:

• Cardiology
• Medical Imaging
• Computational Biology

Background:

• Automated differentiation of wide complex tachycardia (WCT) into ventricular tachycardia (VT) and supraventricular wide complex tachycardia (SWCT) is crucial for patient management.
• Current methods may be improved by quantifying electrical vector changes using vectorcardiogram (VCG) leads.
• The efficacy of VCG lead analysis for automated WCT classification remains to be determined.

Purpose of the Study:

• To investigate whether quantifying mean electrical vector changes within orthogonal VCG leads (X, Y, and Z) can enhance automated VT and SWCT classification.
• To develop and validate a novel VCG-based model for WCT differentiation.

Main Methods:

• A derivation cohort of paired WCT and baseline ECGs was used to develop five logistic regression models, including a novel VCG Model.
• Previously established models (WCT Formula, VT Prediction Model, WCT Formula II) and a Hybrid Model were also included for comparison.
• A separate validation cohort was used to assess the diagnostic performance of all derived models.

Main Results:

• The novel VCG Model, incorporating WCT QRS duration, baseline QRS duration, and amplitude changes in X, Y, and Z leads, demonstrated high diagnostic accuracy (AUC 0.94) in the derivation cohort.
• In the validation cohort, the VCG Model's performance (AUC 0.94) was comparable to existing models, including the WCT Formula (AUC 0.95) and Hybrid Model (AUC 0.95).

Conclusions:

• Custom calculations derived from synthesized VCG signals offer an effective method for automatic WCT differentiation.
• The VCG Model shows potential for improving the accuracy of automated diagnosis of wide complex tachycardias.