RESUMEN
Cardiac arrhythmias cause depolarization waves to conduct unevenly on the myocardial surface, potentially delaying local components with respect to a previous beat when stimulated at faster frequencies. Despite the diagnostic value of localizing the distinct local electrocardiogram (EGM) components for identifying regions with decrement-evoked potentials (DEEPs), current software solutions do not perform automatic signal quantification. Electrophysiologists must manually measure distances on the EGM signals to assess the existence of DEEPs during pacing or extra-stimuli protocols. In this work, we present a deep learning (DL)-based algorithm to identify decrement in atrial components (measured in the coronary sinus) with respect to their ventricular counterparts from EGM signals, for disambiguating between accessory pathways (APs) and atrioventricular re-entrant tachycardias (AVRTs). Several U-Net and W-Net neural networks with different configurations were trained on a private dataset of signals from the coronary sinus (312 EGM recordings from 77 patients who underwent AP or AVRT ablation). A second, separate dataset was annotated for clinical validation, with clinical labels associated to EGM fragments in which decremental conduction was elucidated. To alleviate data scarcity, a synthetic data augmentation method was developed for generating EGM recordings. Moreover, two novel loss functions were developed to minimize false negatives and delineation errors. Finally, the addition of self-attention mechanisms and their effect on model performance was explored. The best performing model was a W-Net model with 6 levels, optimized solely with the Dice loss. The model obtained precisions of 91.28%, 77.78% and of 100.0%, and recalls of 94.86%, 95.25% and 100.0% for localizing local field, far field activations, and extra-stimuli, respectively. The clinical validation model demonstrated good overall agreement with respect to the evaluation of decremental properties. When compared to the criteria of electrophysiologists, the automatic exclusion step reached a sensitivity of 87.06% and a specificity of 97.03%. Out of the non-excluded signals, a sensitivity of 96.77% and a specificity of 95.24% was obtained for classifying them into decremental and non-decremental potentials. Current results show great promise while being, to the best of our knowledge, the first tool in the literature allowing the delineation of all local components present in an EGM recording. This is of capital importance at advancing processing for cardiac electrophysiological procedures and reducing intervention times, as many diagnosis procedures are performed by comparing segments or late potentials in subsequent cardiac cycles.
RESUMEN
AIMS: Slow conduction (SC) anatomical isthmuses (AIs) are the dominant substrate for monomorphic ventricular tachycardia (VT) in patients with repaired tetralogy of Fallot (rTF). This study aimed to evaluate the utility of automated propagational analysis for the identification of SC-AI in patients with rTF. METHODS AND RESULTS: Consecutive rTF patients undergoing VT substrate characterization were included. Automated isochronal late activation maps (ILAM) were obtained with multielectrode HD Grid Catheter. Identified deceleration zones (DZs) were compared with both SC-AI defined by conduction velocity (CV) (<0.5â m/s) and isthmuses of induced VT for mechanistic correlation. Fourteen patients were included (age 48; p25-75 35-52 years; 57% male), 2 with spontaneous VT and 12 for risk stratification. Nine VTs were inducible in seven patients. Procedure time was 140 (p25-75 133-180) min and mapping time 29.5 (p25-75 20-37.7) min, using a median of 2167 points. All the patients had at least one AI by substrate mapping, identifying a total of 27 (11 SC-AIs). Isochronal late activation maps detected 10 DZs mostly in the AI between ventricular septal defect and pulmonary valve (80%). Five patients had no DZs. A significant negative correlation between number of isochrones/cm and CV was observed (rho -0.87; P < 0.001). Deceleration zones correctly identified SC-AI (90% sensitivity; 100% specificity; 0.94 accuracy) and was related to VT inducibility (P = 0.006). Deceleration zones co-localized to the critical isthmus of induced VTs in 88% of cases. No complications were observed. CONCLUSION: Deceleration zones displayed by ILAM during sinus rhythm accurately identify SC-AIs in rTF patients allowing a safe and short-time VT substrate characterization procedure.
Asunto(s)
Ablación por Catéter , Válvula Pulmonar , Taquicardia Ventricular , Tetralogía de Fallot , Humanos , Masculino , Persona de Mediana Edad , Femenino , Tetralogía de Fallot/cirugía , Frecuencia Cardíaca/fisiología , Arritmias Cardíacas , Ablación por Catéter/efectos adversosRESUMEN
No disponible