Using Association Rules to Obtain Sets of Prevalent Symptoms throughout the COVID-19 Pandemic: An Analysis of Similarities between Cases of COVID-19 and Unspecified SARS in São Paulo-Brazil

The efficient recognition of symptoms in viral infections holds promise for swift and precise diagnosis, thus mitigating health implications and the potential recurrence of infections. COVID-19 presents unique challenges due to various factors influencing diagnosis, especially regarding disease symp...

Full description

Saved in:
Bibliographic Details
Published in:International journal of environmental research and public health 2024-09, Vol.21 (9), p.1164
Main Authors: Marques, Julliana Gonçalves, Carvalho, Bruno Motta de, Guedes, Luiz Affonso, Da Costa-Abreu, Márjory
Format: Article
Language:English
Subjects:
Algorithms
Brazil - epidemiology
Cardiovascular disease
COVID-19
COVID-19 - epidemiology
Datasets
Diabetes
Diarrhea
Dyspnea
Experiments
Fever
Humans
Hypertension
Infections
Machine learning
Mortality
Olfaction disorders
Pandemics
Patients
Prevalence
Respiratory distress syndrome
SARS-CoV-2
Severe Acute Respiratory Syndrome - epidemiology
Severe acute respiratory syndrome coronavirus 2
Taste disorders
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The efficient recognition of symptoms in viral infections holds promise for swift and precise diagnosis, thus mitigating health implications and the potential recurrence of infections. COVID-19 presents unique challenges due to various factors influencing diagnosis, especially regarding disease symptoms that closely resemble those of other viral diseases, including other strains of SARS, thus impacting the identification of useful and meaningful symptom patterns as they emerge in infections. Therefore, this study proposes an association rule mining approach, utilising the Apriori algorithm to analyse the similarities between individuals with confirmed SARS-CoV-2 diagnosis and those with unspecified SARS diagnosis. The objective is to investigate, through symptom rules, the presence of COVID-19 patterns among individuals initially not diagnosed with the disease. Experiments were conducted using cases from Brazilian SARS datasets for São Paulo State. Initially, reporting percentage similarities of symptoms in both groups were analysed. Subsequently, the top ten rules from each group were compared. Finally, a search for the top five most frequently occurring positive rules among the unspecified ones, and vice versa, was conducted to identify identical rules, with a particular focus on the presence of positive rules among the rules of individuals initially diagnosed with unspecified SARS.
ISSN:1660-4601
1661-7827
1660-4601
DOI:10.3390/ijerph21091164