ArticleBMC medical informatics and decision making2025
Development of an explainable machine learning model for predicting device-related pressure injuries in clinical settings.
Article in BMC medical informatics and decision making, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 2 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
2 citing papers in PubMed.
- Exploratory Analysis of Factors Associated With Pressure Injuries in Surgical Patients Assessed With the ELPO Scale.International journal of nursing practice · 2026Article
- Impact of sense of coherence and social support on disease activity in Crohn's disease: a network analysis.Frontiers in public health · 2025Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
8 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
backgroundDevice-related pressure injury (DRPI) is a prevalent and severe problem for patients using medical devices. Timely identification of patients at high risk of DRPI is crucial for healthcare providers to make informed decisions and prevent DRPI quickly. Given the rapid advancements in computer technology, we aimed to develop an interpretable artificial intelligence (AI) model for predicting DRPI, utilizing SHAP (SHapley Additive exPlanations) to enhance the model's transparency and provide insights into feature importance.
methodsWe enrolled 675 study subjects (225 in the DRPI group and 450 in the non-DRPI group) from a single medical center between January 2019 and December 2020. Python was used to perform classification models, including extreme gradient boosting (XGBoost), random forest (RF), decision tree (DT), Logistic Regression (LR), support vector machine (SVM), and K-Nearest neighbors (KNN). We evaluated the performance of the six models using area under the ROC curve (AUC), specificity, accuracy, and sensitivity, with the dataset split into a 80% training set and a 20% testing set. We utilized several analyses, such as SHAP and Uniform Manifold Approximation and Projection (UMAP), to explore the potential contribution of different characteristics in our risk prediction models.
resultsIn the test set, XGBoost model outperformed the other models (AUC = 0.964). The interpretation of the model using SHAPscores revealed that the length of stay, instrument type, emergency admissions, instrument material, and instrument duration of use are the top five most important features in predicting DRPI.
conclusionOur study demonstrated that the development of DRPI in patients can be accurately predicted using the machine learning (ML) model. The findings not only provide clinical caregivers with a valuable framework to identify patients at high risk of DRPI, but also lay the groundwork for developing targeted preventive strategies and personalized interventions.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.