A carregar...
Publicação
Pressure ulcer prediction in intensive care units using artificial intelligence: a scoping review
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 45.42 KB | Adobe PDF |
Orientador(es)
Resumo(s)
Aim: Pressure ulcers significantly challenge healthcare, affecting quality of life and straining systems. Despite most being preventable, these injuries remain common in Intensive Care Units (ICUs). Effective risk identification is crucial, but traditional scales have limitations, prompting the development of new tools. Artificial intelligence (AI) offers a promising approach to dynamically identify and prevent pressure ulcers in critical contexts. This review assesses the literature on AI technologies for predicting pressure ulcers in critically ill patients in ICUs, identifying knowledge gaps and guiding future research. Method: Using the Joanna Briggs Institute’s scoping review methodology, this study focuses on AI technologies for pressure ulcer prediction (C) in critically ill patients (P) admitted to ICUs (C). The protocol was previously registered on the Open Science Framework. Search was conducted across relevant electronic databases yielding 137 publications. Results / Discussion: Fourteen studies were included. Most used cohort designs with electronic health records to train machine learning algorithms. AI models were trained using 6 to 86 variables, with the best models showing Area Under the Receiver Operating Characteristic Curve (AUROC) values ranging from 0,73 to 0,99, reflecting high predictive accuracy. One publication studied the impact of an AI model in clinical practice, achieving a reduction in pressure ulcer prevalence from 21,26% to 4,04% and a decrease in ICU length of stay from 7,63 to 5,17 days, demonstrating high adherence and significant clinical impact. Conclusion: AI technologies offer a dynamic solution to improve the timely prediction of pressure ulcers, addressing limitations of traditional tools. This review synthesizes current findings and directs future research toward enhancing ICU care.
Descrição
Palavras-chave
Contexto Educativo
Citação
Alves, J., Azevedo, R., Marques, A., & Alves, P. (2025). Pressure ulcer prediction in intensive care units using artificial intelligence: A scoping review. 277-277. Abstract from 35th Conference of the European Wound Management Association, Barcelona, Spain.