Quantifying ECG Redundancy through Mutual Information Analysis among Leads and Its Application in CNNs
Elisa Ramírez, Samuel Ruiperez-Campillo, Raúl Alós, Francisco Castells, Rubén Casado-Arroyo, José Millet · Computing in cardiology · 2024
The electrocardiogram (ECG) is widely used in clinical practice for diagnosing diseases.However, conventional ECGs contain 12 projections of the three-dimensional cardiac dipole vector, resulting in significant redundancy which may negatively impact artificial intelligence algorithms performance.To address this issue, we developed a metric based on mutual information to quantify redundancy in standard ECGs.We used two strategies for redundancy reduction: eliminating leads and applying linear transformations to the original ECG leads.We employed a convolutional neural network (CNN) with inputs generated from redundancy elimination.We found that reducing the input to three orthogonal threedimensional coordinates had minimal impact on the model's performance owing its slight decrease in performance to potential elimination of relevant information on the ECG wave morphologies.However, using six channels mitigated distortion by minimizing information loss, even with increased redundancy.Clinical Relevance -This study establishes an objective criterion for selecting cardiac vector projections that minimize redundancy while maintaining diagnostic capacity in the CNN.This serves as an objective and robust criterion for selecting the most informative projections in wearables or Holter monitors.