Gaetano Caruso
Automated Detection of ECG Electrode Interchange Using Machine Learning.
Rel. Vincenzo Randazzo. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Informatica (Computer Engineering), 2026
|
Preview |
PDF (Tesi_di_laurea)
- Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives. Download (13MB) | Preview |
Abstract
This thesis investigates the development of a deep learning pipeline for the detection of electrode interchange in 12-lead electrocardiograms (ECGs). Recognizing the incorrect placement of the electrodes is vital, since electrode interchange can significantly alter signal morphology, potentially mimicking or concealing clinically relevant cardiac conditions. Limb and precordial leads were treated as two separate problems with corresponding separate pipelines. At first, the ECG undergoes a signal preprocessing pipeline, which includes power-line interference removal, band-pass filtering, resampling, signal quality assessment, robust normalization, and segmentation into two-second windows. Limb electrode interchange was formulated as a six class classification task, including the normal configuration and five interchange configurations.
For the precordial interchange, sixteen classes were used, including the normal configuration and fifteen pairwise exchanges among leads V1-V6
Relatori
Anno Accademico
Tipo di pubblicazione
Numero di pagine
Corso di laurea
Classe di laurea
URI
![]() |
Modifica (riservato agli operatori) |
