Previously submitted to: JMIR Medical Informatics (no longer under consideration since Nov 20, 2025)
Date Submitted: Sep 22, 2025
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
AI-based Mobile Application for Pressure Injury Healing: Real-World Evidence from a Retrospective Study
ABSTRACT
Background:
Pressure injuries (PIs) are a major concern in hospital care, contributing to prolonged hospital stays, reduced quality of life, and increased healthcare costs. Accurate and timely staging is critical for effective management, highlighting the need for practical and clinically validated tools that can support precise PI assessment and improve patient outcomes.
Objective:
This study aimed to evaluate both the technical performance and the real-world clinical effectiveness of an artificial intelligence (AI)-based application for PI staging and dressing selection in a tertiary care hospital setting.
Methods:
We developed a convolutional neural network–based mobile application using 9,083 wound images and conducted external validation with 448 independent images. The application was embedded into barcode point-of-care (BPOC) devices for bedside use and real-time electronic medical record (EMR) integration. To assess clinical effectiveness, we conducted a retrospective analysis of 610 matched cases using propensity score matching, comparing changes in Pressure Ulcer Scale for Healing (PUSH) scores over 7 days between the app-use group and the non-app-use group. Subgroup analyses were performed according to the timing of app usage and PI stage.
Results:
The model achieved high diagnostic accuracy across PI stages, with a weighted area under the receiver operating characteristic curve (AUROC) of 0.969 in internal testing and 0.971 in external validation. In the clinical evaluation, the use of the application was associated with significantly greater improvement in PUSH scores for Stage 2 and above PIs (48.80% vs. 35.58%, OR=1.825, 95% CI=1.152–2.891, P=.010). Subgroup analysis revealed that early app usage led to higher odds of healing improvement (51.69% vs. 36.08%, adjusted OR=1.862, 95% CI=1.004–3.454, P=.048), while delayed usage showed a positive trend without statistical significance.
Conclusions:
This study demonstrates that an AI-based application can be effectively deployed at the bedside to improve clinical decision-making and enhance pressure injury healing, particularly when implemented during the initial phase of pressure injury management. To our knowledge, this is the first real-world study to demonstrate both technical validity and clinical benefits of AI -powered PI staging and dressing recommendations.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.