Skip to main navigation Skip to search Skip to main content

Ferramentas de machine learning na gestão da doença crónica: scoping review

Translated title of the contribution: Machine learning tools in chronic disease management: scoping review

Research output: Contribution to journalReview articlepeer-review

Abstract

Background: the implementation of technologies based on Artificial Intelligence (AI) in the health sector, in particular machine learning (ML), has had a significant transformational effect. Their use improves disease prediction, classification and diagnosis, benefiting both users and healthcare professionals. Objective: to map ML tools for chronic disease management, with relevance to nursing care for people with chronic diseases. Methodology: scoping review based on the recommendations of the Joanna Briggs Institute. The MEDLINE Complete via PUBMED, CINAHL Complete via EBSCO, SCOPUS, OpenGrey, RCAAP and DART-Europe databases were used, with no time limit. Results: seven articles were included and 9 AI tools associated with chronic disease management were identified, namely chronic kidney disease, chronic obstructive pulmonary disease, hepatitis C, heart failure and chronic venous insufficiency. Conclusion: the tools identified have the potential to contribute to improving nursing care, particularly in identifying risk factors associated with chronic diseases, detecting exacerbations early, continuously monitoring and evaluating the effectiveness of treatment and supporting clinical decision-making.
Translated title of the contributionMachine learning tools in chronic disease management: scoping review
Original languagePortuguese
Pages (from-to)1-12
Number of pages12
JournalRevista de Investigação & Inovação em Saúde
Volume7
Issue number1
DOIs
Publication statusPublished - 2024
Externally publishedYes

Keywords

  • Machine learning
  • Chronic disease
  • Nursing

Fingerprint

Dive into the research topics of 'Machine learning tools in chronic disease management: scoping review'. Together they form a unique fingerprint.

Cite this