Banca de QUALIFICAÇÃO: ANGÉLICA DRIELLY SANTOS DE QUADROS

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
DISCENTE : ANGÉLICA DRIELLY SANTOS DE QUADROS
DATA : 26/06/2025
HORA: 10:00
LOCAL: Sala 107 Bloco Zeta Campus São Bernardo do Campo
TÍTULO:

Animal model for non-invasive electrocardiographic imaging during cardiac rhythms

PÁGINAS: 80
RESUMO:

Cardiovascular diseases remain the leading cause of death worldwide, accounting for millions of fatalities annually. Among these, cardiac arrhythmias pose a significant challenge in clinical electrophysiology, necessitating advanced, non-invasive mapping techniques for accurate diagnosis and treatment planning. Electrocardiographic imaging (ECGi) has emerged as a promising solution by reconstructing epicardial electrical activity from body surface potential mapping (BSPM) and three-dimensional heart-torso geometries. However, ECGi validation remains difficult due to the challenges of acquiring direct epicardial data for comparison.

This study aims to develop a customized ECGi approach for generating epicardial electrophysiological maps in an in situ experimental rabbit heart model. Using a hexagonal torso-tank filled with a conductive medium, BSPM signals will be recorded and processed to reconstruct epicardial activation maps. The study investigates both normal sinus rhythm and induced arrhythmias, optimizing signal preprocessing techniques to enhance reconstruction accuracy.

The research evaluates ECGi accuracy by comparing reconstructed epicardial maps against both optical and electrical recordings, with optical mapping serving as the gold-standard reference. Two established regularization techniques - zero-order Tikhonov and Truncated Singular Value Decomposition (TSVD) at various orders - were implemented, both with L-curve-based parameter selection to enhance reconstruction stability and fidelity. In parallel, a prototype deep-learning framework - featuring 3D-convolutional encoders coupled with LSTM layers and trained on the same experimental dataset – is under development to capture complex spatiotemporal features. Although still preliminary, this AI-driven approach will be benchmarked against the classic methods to assess its potential for future ECGi enhancement.

Key metrics such as root mean square error (RMSE), cross-correlation, local activation time (LAT), and frequency-domain analysis will be used for performance assessment. This proof-of-concept study seeks to refine ECGi methodologies, address the inverse problem’s inherent challenges, and contribute to the advancement of non-invasive cardiac mapping techniques.

MEMBROS DA BANCA:
Presidente - Interno ao Programa - 1188948 - JOAO LAMEU DA SILVA JUNIOR
Membro Titular - Examinador(a) Externo à Instituição - MARIA DE LA SALUD GUILLEM SANCHEZ - UPV
Membro Titular - Examinador(a) Externo à Instituição - ÓSCAR BARQUERO PÉREZ - URJC
Membro Suplente - Examinador(a) Interno ao Programa - 2123676 - OLAVO LUPPI SILVA
Membro Suplente - Examinador(a) Externo à Instituição - RENATA VALERI DE FREITAS - USP
Membro Suplente - Examinador(a) Externo à Instituição - MATHEUS CARDOSO MORAES - UNIFESP