Defesa de Proposta de Dissertação: Sara Alexandre Fonsêca
Data da publicação: 7 de novembro de 2024 Categoria: Notícias, Proposta de Dissertação
Título: Anomaly Detection in ECG Signals for Arrhythmia Identification Using Machine Learning Techniques
Data: 14/11/2024
Horário: 16h00
Local: Sala de Seminários – Bloco 952
Resumo:
Anomaly detection in Electrocardiogram (ECG) data is crucial for the early diagnosis and treatment of cardiac conditions, with a specific focus on arrhythmia identification. This work explores machine learning and deep learning techniques to detect anomalies in ECG signals within a predominantly unsupervised learning framework. The methodology employs two main approaches: (i) feature engineering from ECG exams; and (ii) feature learning directly from ECG time-series signals through deep techniques. The ECG data used is derived from the publicly available MIT-BIH Arrhythmia Database, providing standardized and widely accepted records for evaluating arrhythmia analysis algorithms. This approach not only ensures accessibility and reproducibility of the results but also highlights the effectiveness of unsupervised learning for automatic feature extraction from temporal signals. The findings indicate a significant potential for the development of AI-assisted diagnostic tools, advancing the monitoring and precise diagnosis of cardiac arrhythmias.
Banca examinadora:
- Prof. Dr. João Paulo do Vale Madeiro (MDCC/UFC – Orientador)
- Prof. Dr. César Lincoln Cavalcante Mattos (MDCC/UFC – Coorientador)
- Prof. Dr. Luís Otávio Rigo Júnior (UNILAB)
