Pressure Injury
Conditions
Keywords
Artificial intelligence, evidence-based, intensive care, machine learning
Brief summary
This study aims to develop a mobile nursing care system that uses machine learning to support pressure injury staging and to evaluate its effect on pressure injury healing in an intensive care unit. The system will provide stage-based care guidance and support wound monitoring. The clinical evaluation will use a quasi-experimental, pretest-posttest design with a control group. A total of 70 patients with pressure injuries are planned to be included: 35 in the intervention group and 35 in the control group. Data will be collected using a patient information and monitoring form, the Braden Scale, and the Bates-Jensen Wound Assessment Tool. Changes in wound condition and healing will be compared between the groups.
Detailed description
This study consists of two stages: First, a machine learning-based system will be developed to classify pressure ulcers based on clinical images, measure wound surface area, and provide stage-specific, evidence-based nursing care guidelines. Secondly, the system will be evaluated in an intensive care unit through a quasi-experimental study. During the control phase, pressure injury assessment and nursing care will be carried out in accordance with current practices. Subsequently, nurses will receive training on the use of the mobile system. In the intervention phase, nurses will use the system to support the developed processes for pressure injury staging, care planning, and wound monitoring. Patient characteristics, wound characteristics, and the healing process will be recorded using study forms and assessment tools. Changes in wound status will be evaluated throughout the observation period and compared between the control and intervention groups. The study will demonstrate the effect of the developed system on the healing of pressure injuries.
Interventions
DEVICEMobile Application for Pressure Injury Staging and Nursing Care Guidance
Specially trained intensive care nurses will use a mobile application during the assessment and care of patients with pressure injuries. The application employs a machine learning model to suggest the stage of the pressure injury based on an image of the wound and provides evidence-based, stage-specific nursing care guidance. It also enables objective measurement of the wound surface area. The application is intended for use alongside standard clinical care, and clinical decision-making will remain the responsibility of the healthcare team.
Patients in this arm will receive standard pressure injury care provided in the intensive care unit according to existing clinical practice. Nurses will assess and document the wounds without using the study's mobile application. Pressure injury outcomes will be measured using the same study assessment procedures as in the intervention arm.
Sponsors
Ertugrul Sari
The Scientific and Technological Research Council of Turkey
Study design
Allocation
NON_RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
NONE
Intervention model description
Two groups of patients with pressure injuries will be compared. The control group will receive standard clinical care. The intervention group will receive care supported by a mobile application that provides pressure injury staging and evidence-based, stage-specific nursing care guidance. Control group data will be collected before implementation of the mobile application, followed by intervention group data collection.
Eligibility
Sex/Gender
ALL
Age
18 Years to 65 Years
Healthy volunteers
No
Inclusion criteria
* Aged 18 to 65 years. * Diagnosed with a pressure injury. * Hemodynamically stable, as assessed by the treating clinician. * Remained in the intensive care unit for at least 48 hours after identification of the pressure injury. * The patient or their legal representative provides verbal and written informed consent after receiving information about the study.
Exclusion criteria
* Intensive care unit stay of less than 48 hours after identification of the pressure injury. * The wound cannot be photographed because of extensive infection or necrotic tissue in the wound area. * The patient or their legal representative does not consent to participation.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Change in Bates-Jensen Wound Assessment Tool Score | Days 1, 3, 5, 7, 9, 11, and 13 | Pressure injury status will be assessed using the Bates-Jensen Wound Assessment Tool. Scores range from 13 to 65, with lower scores indicating better wound status. Change from Day 1 to Day 13 and scores at each assessment will be compared between groups. |
| Change in Pressure Injury Surface Area | Days 1, 3, 5, 7, 9, 11, and 13 | Pressure injury surface area will be recorded in cm² on Days 1, 3, 5, 7, 9, 11, and 13. In the intervention group, the mobile application will calculate the wound area. Changes from Day 1 to Day 13 and trends over time will be compared between groups. A decrease in surface area will indicate wound healing. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Change in Braden Scale Score | Days 1, 3, 5, 7, 9, 11, and 13 | Risk of pressure injury will be assessed using the Braden Scale at each assessment. Scores range from 6 to 23, with lower scores indicating greater risk. Change from Day 1 to Day 13 and scores over time will be compared between groups. |
Countries
Turkey (Türkiye)
Outcome results
None listed