Pattern Recognition of Electrocardiogram Signals through Deep Learning and Recurrence Analysis

• Grant number: 24/10715-1
• Support Opportunities: Scholarships in Brazil - Master
• Start date: January 01, 2025
• End date: April 30, 2026
• Field of knowledge: Engineering - Electrical Engineering - Telecommunications
• Principal Investigator: Levy Boccato
• Grantee: João Pedro de Oliveira Pagnan
• Host Institution: Faculdade de Engenharia Elétrica e de Computação (FEEC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Company: Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Elétrica e de Computação (FEEC)
• Associated research grant: 20/09838-0 - BI0S - Brazilian Institute of Data Science, AP.PCPE

Abstract

This master's project focuses on the study of pattern recognition techniques in the context of electrocardiogram (ECG) signals, with a primary emphasis on detecting cardiovascular diseases (CVD). CVDs are the leading cause of death worldwide, with a significant portion of these deaths occurring in low and middle-income countries.The pattern recognition technique to be studied in the proposed project involves the use of deep learning (DL) applied to recurrence plots (RP), distance matrices (DP), and quantitative recurrence analysis (RQA) metrics of ECG signals. This approach is chosen due to unexplored factors in existing research employing these ECG signal analysis techniques.The objectives of this project include: (i) implementing RP, DP, and RQA metrics in a Python coding environment; (ii) conducting sensitivity tests of ECG signal classification by DL models to the parameters of these metrics; (iii) analyzing the interpretability of classification performed by the model based on RP, DP, and RQA; (iv) performing a comparative analysis of classification using DP versus RP and RQA.This set of objectives establishes a robust framework for the master's research, characterized by its potential to generate original contributions and its interdisciplinary and formative nature, which provides a wide range of possibilities for subsequent doctoral research.