Using AI Within Digital Wound Monitoring of Surgical Wounds to Prioritise Non-healing Wounds for Urgent Review (WISDOM)

Wound Imaging Software and Digital platfOrM to Detect and Prioritise Non-healing Surgical Wounds (WISDOM)

The goal of this clinical trial is to learn if artificial intelligence within digital wound monitoring of surgical wounds can prioritise non-healing wounds for urgent review in patients having first/redo CABG surgeries with or without additional cardiac procedures; could include any of the following: any gender, age ≥18. The main question it aims to answer is:

Can investigators successfully develop artificial intelligence to prioritise images of patients' surgical wounds that are failing to heal or are infected, in order to facilitate early treatment?

Researchers will compare the wound prioritisation module with standard care to determine safety and acceptability outcomes.

Participants from the control group will:

• Have standard post-operative wound care follow-up at day 30 (survey, interview and phone call)
• Have standard post-operative wound care follow-up at day 60 (phone call)

Participants from the intervention group will:

• Use platform with new AI prioritisation module for 30 days after surgery
• Be contacted via SMS text message seven days, fourteen days and twenty-one days after surgery with the link request remaining open for 6 days until the next request is sent out
• Submit a photo of their wound and complete the UKHSA wound surveillance questionnaire
• Have standard post-operative wound care follow-up at day 30 (survey, interview and phone call)
• Have standard post-operative wound care follow-up at day 60 (phone call)

Study Overview

• Status: Completed
• Conditions: Surgical Wound; Heart; Surgery, Heart, Functional Disturbance as Result; Wound Healing Delayed
• Intervention / Treatment: Device: Isla wound prioritisation module; Other: Standard follow-up

Detailed Description

Background and study aims Over 10 million surgical operations are performed in England annually with approximately 2.1 million having problems with wound healing, of which 500,000 lead to infection. Most of these wound problems happen after patients have been discharged from the hospital. They need to be identified and treated early to prevent the problem from worsening.

Digital remote surgical wound monitoring is beginning to be used to monitor patients' surgical wounds at home after discharge from the hospital. This offers regular assessment when wound problems are most likely to develop. Early evaluations of digital wound monitoring suggest it improves clinical outcomes and has high patient satisfaction; however, it creates a new additional workload for clinicians.

Who can participate? Patients ≥18 years old having first/redo coronary artery bypass graft (CABG) surgeries with or without adjunct cardiac procedures such as valve replacement, or chest reopening during same admission as index surgery, and either no infection, or an existing non-infected wound complication, or any other infection except surgical site, at any of two recruitment sites (St Bartholomew's Hospital, London and Freeman Hospital, Newcastle). Patients without a smartphone/with physical disability/with visual impairment will be eligible if they are willing to use a smartphone or internet provided by the study, or their next of kin or carer is able-bodied or has a smartphone.

What does the study involve? The study will assess a new component for a digital wound monitoring platform, which has been developed and has recently received HRA approval to be validated for predictivity, sensitivity and specificity, and inter-rater reliability. The new component uses artificial intelligence (AI) to identify 'red flags' on the images patients submit to the wound monitoring platform. Images that have a possible red flag are then identified for urgent priority review. This helps clinicians manage this new workload by allowing the most urgent cases to be reviewed first.

A total of 120 patients in two hospitals will be invited to take part in the study. All participants (patients who take part) will receive normal wound care follow-up after surgery, and half of the participants will also receive the digital wound monitoring system with the AI to identify wounds which need urgent assessment.

What are the possible benefits and risks of participating? Participants allocated to the digital wound monitoring with AI may benefit from regular and ongoing wound assessment. Participants allocated to standard wound care may not receive additional direct benefits. This study will help improve the approach to Wound care.

There are no foreseeable disadvantages involved with taking part since all participants will receive their usual standard wound care follow-up. Being involved will require participants to give some of their own time.

Where is the study run from? This study is a collaboration of researchers and surgeons across the country. It is co-ordinated by Derby Clinical Trials Support Unit, on behalf of the Sponsor, Guys and St Thomas' NHS Foundation Trust (UK)

When is the study starting and how long is it expected to run for? December 2023 to December 2025

Who is funding the study? National Institute for Health and Care Research (NIHR) (UK)

• Study Type: Interventional
• Enrollment (Actual): 120
• Phase: Not Applicable

Contacts and Locations

Study Locations

• United Kingdom
  • London, United Kingdom, EC1A 7BE
    • St Bartholomew's Hospital
  • Newcastle Upon Tyne, United Kingdom, NE7 7DN
    • Freeman Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Patients having first/redo CABG surgeries with or without adjunct cardiac procedures such as valve replacement, or chest reopening during same admission as index surgery, and either no infection, or an existing non-infected wound complication, or any other infection except surgical site.
• Patients without a smartphone/with physical disability/with visual impairment will be eligible if they are willing to use a smartphone or internet provided by the study, or their next of kin or carer is able-bodied or has a smartphone.

Exclusion Criteria:

• Patients having CABG requiring ventricular assist device (VAD) or
• extracorporeal membrane oxygenation (ECMO), or
• ventilated or unconscious patients, or
• pre-existing surgical site infection.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Prevention
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Active Comparator: Isla digital wound monitoring + Standard follow-up; Patients in the intervention group will be contacted via SMS text message seven days, fourteen days and twenty-one days after surgery with the link request remaining open for 6 days until the next request is sent out. The exception being the last request link which will remain open until 30 days after surgery. In the requests patients are asked to submit a photo of their wound and complete the UKHSA wound surveillance questionnaire. Participants can also submit an image during the 30 days whenever they have a concern.	Device: Isla wound prioritisation module; The intervention group will use the artificial intelligence enabled platform with the new wound prioritisation module for 30 days after surgery in addition to standard post-operative wound follow-up care for 60 days after surgery.; Other: Standard follow-up; Standard care, mapped during the economic scoping exercise may include; out-patient appointments, advised to contact GP, or no follow-up.
Active Comparator: Standard follow-up; Patients in the control group will have standard post-operative wound care follow-up for 60 days.	Other: Standard follow-up; Standard care, mapped during the economic scoping exercise may include; out-patient appointments, advised to contact GP, or no follow-up.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Quality of images assessed by clinicians, in frequencies	A quality image is one that can be used to make a clinical decision	30 days
Quality of images assessed by clinicians, in percentages	A quality image is one that can be used to make a clinical decision	30 days
Clinician satisfaction using surveys	Acceptability of the intervention including attitude, burden, perceived effectiveness, ethicality, intervention coherence, opportunity costs and self-efficacy. Will also collect acceptability of being involved as a study participant. Survey variables that are continuous will be reported with means & 95% confidence intervals (95% CI), if shown to be normally distributed, using a normality plot, otherwise will be reported with medians & Interquartile Ranges (IQR). The categorical variables will be reported with frequencies & percentages.	30 days
Clinician satisfaction using interviews	Acceptability of the intervention including attitude, burden, perceived effectiveness, ethicality, intervention coherence, opportunity costs and self-efficacy. Will also collect acceptability of being involved as a study participant. Survey variables that are continuous will be reported with means & 95% confidence intervals (95% CI), if shown to be normally distributed, using a normality plot, otherwise will be reported with medians & Interquartile Ranges (IQR). The categorical variables will be reported with frequencies & percentages.	30 days
Patient satisfaction using surveys	Acceptability of the intervention including attitude, burden, perceived effectiveness, ethicality, intervention coherence, opportunity costs and self-efficacy. Will also collect acceptability of being involved as a study participant. Survey variables that are continuous will be reported with means & 95% confidence intervals (95% CI), if shown to be normally distributed, using a normality plot, otherwise will be reported with medians & Interquartile Ranges (IQR). The categorical variables will be reported with frequencies & percentages.	30 days
Patient satisfaction using interviews	Acceptability of the intervention including attitude, burden, perceived effectiveness, ethicality, intervention coherence, opportunity costs and self-efficacy. Will also collect acceptability of being involved as a study participant. Survey variables that are continuous will be reported with means & 95% confidence intervals (95% CI), if shown to be normally distributed, using a normality plot, otherwise will be reported with medians & Interquartile Ranges (IQR). The categorical variables will be reported with frequencies & percentages.	30 days
Recruitment rate (number and percentage of eligible patients recruited to the study). Patients who were consented but not randomised will not be included in the recruitment rate.	Recruitment rate (including access and barriers to recruitment and willingness to be randomised) will be assessed as a percentage of all eligible patients, regardless of treatment allocation.	30 days
Adherence with the module (intervention group only). Adherence will be reported as the number and percentage of adherent patients in the intervention group.	To be adherent a patient needs to submit 1 photo within the 30 day period.	30 days
Loss to follow-up (number and percentage of intervention patients complete the study)	Loss to follow-up will be assessed based on all randomised patients.	30 days
Number of wound problems/infections (in number and percentage)	Analysis will be reported based on allocated treatment.	30 days
Severity of wound problems/infections	Analysis will be reported based on allocated treatment.	30 days
Wound-related hospital admissions (in number and percentage)	Analysis will be reported based on allocated treatment.	30 days
Prescribed antibiotics (in number and percentage)	Analysis will be reported based on allocated treatment.	30 days
Time to review images in minutes (intervention only)	Analysis will be reported based on allocated treatment.	30 days
Further surgery to treat wounds in number and percentage	Analysis will be reported based on allocated treatment.	30 days
Prescribed wound treatments	Analysis will be reported based on allocated treatment.	30 days
Number of clinic visits in number and percentage	Analysis will be reported based on allocated treatment.	30 days
Number of General Practice visits in frequency and percentage	Analysis will be reported based on allocated treatment.	30 days
Patient travel time in minutes		30 days
Quality of life survey using SF-6D v2		At baseline, 30 days, and 60 days

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Reasons for compliance/non-compliance will be explored in the staff surveys	Survey data will be analysed using descriptive statistics, and qualitative data from the interviews will be analysed using thematic analysis.	60 days
Reasons for compliance/non-compliance will be explored in the staff interviews	Survey data will be analysed using descriptive statistics, and qualitative data from the interviews will be analysed using thematic analysis.	60 days
Reasons for compliance/non-compliance will be explored in the patient intervention group surveys	Survey data will be analysed using descriptive statistics, and qualitative data from the interviews will be analysed using thematic analysis.	60 days
Reasons for compliance/non-compliance will be explored in the patient intervention group interviews	Survey data will be analysed using descriptive statistics, and qualitative data from the interviews will be analysed using thematic analysis.	60 days
Feasibility outcome: access/barriers to participation using survey		60 days
Feasibility outcome: access/barriers to participation using interviews		60 days
Feasibility outcome: willingness of participants to be randomised in number and percentage		60 days
Feasibility outcome: attrition rate in percentage		60 days
Feasibility outcome: time (in minutes) to conduct telephone assessments to patients and phone calls to GPs to collect antibiotic data		60 days
Feasibility outcome: resources to conduct telephone assessments to patients and phone calls to GPs to collect antibiotic data		60 days
Number of photos received per patient (number and percentage)		60 days
Number of wound images/non wound images (number and percentage)		60 days
Number of requests (for images) complied with - intervention group only (number and percentage)		60 days
Number photos initiated by patients - intervention group only (number and percentage)		60 days
Number of follow-up requests (number and percentage)		60 days

Collaborators and Investigators

• Sponsor: Guy's and St Thomas' NHS Foundation Trust
• Collaborators: University of Nottingham; Liverpool Heart and Chest Hospital NHS Foundation Trust; National Institute for Health Research, United Kingdom; Central and North West London NHS Foundation Trust; Islacare Ltd; Derby Clinical Trials Unit; East Midlands Research Support Service; Centre for Healthcare Equipment and Technology Adoption; Health Innovation East Midlands
• Investigators: Principal Investigator: Alex Shipolini, Barts & The London NHS Trust; Principal Investigator: Faruk Oezalp, Newcastle Hospitals NHS Foundation Trust

Study record dates

Study Major Dates

• Study Start (Actual): August 24, 2024
• Primary Completion (Actual): March 23, 2025
• Study Completion (Actual): March 23, 2025

Study Registration Dates

• First Submitted: April 29, 2024
• First Submitted That Met QC Criteria: June 20, 2024
• First Posted (Actual): June 26, 2024

Study Record Updates

• Last Update Posted (Actual): June 12, 2025
• Last Update Submitted That Met QC Criteria: June 11, 2025
• Last Verified: June 1, 2025

More Information

Terms related to this study

• Keywords: surgery; artificial intelligence; surgical site infection; infection control; wound management
• Additional Relevant MeSH Terms: Surgical Wound; Wounds and Injuries
• Other Study ID Numbers: UHDB/2022/024

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: On completion of the study, anonymised quantitative data from the surveys, interviews and phone calls and anonymised themed analysis of qualitative data from the interviews will be stored in and made publicly available through the University of Nottingham data repository (https://rdmc.nottingham.ac.uk/).

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No