Finding patterns in cardiologic diseases using a data-driven approach

• Autor(a) principal: Gomes, Filipa Isabel Ribeiro
• Data de Publicação: 2023
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/29148
• topic: Data science; Cardiologia -- Cardiology; Doença cardiovascular -- Cardiovascular disease; Diagnóstico -- Diagnosis; Alert system; Inteligência artificial -- Artificial intelligence; Análise de dados -- Data analysis; Sistema de alertas

Globally, cardiovascular disease (CD) is the leading cause of death. Several guidelines for the treatment of CD have been published with the aim of improving the quality of care and reducing costs. Thus, it is increasingly important to detect and diagnose cardiovascular diseases early. This study aims to build an algorithm to predict whether a patient will exceed their heart rate. In addition, the goal was to build an alert system that monitors the patient's clinical status and, whenever there is a change, according to some parameters, the doctor receives a message automatically. This study was based on a set of data from Santa Maria Hospital in Lisbon, obtained through Digital Services Agreements developed under the FCT project DSAIPA/AI/0122/2020 AIMHealth - Artificial Intelligence Based Mobile Applications for Public Health Response. The data-centric method followed the CRISP-DM Data Mining (DM) methodology. Based on the dataset it was possible, following this methodology, to develop a Machine Learning (ML) algorithm that could predict in advance whether the patient would exceed the interquartile range of their heart rate. We found that our ML algorithm was able to predict cardiac problems in 90% of the cases and that our alert system was effective in early detection of cardiac problems in patients. This study has shown that using ML is a valuable tool for detecting the worsening of a patient's health condition.