Trial Review

Technical difficulties have been reported by some users of the search function and is being investigated by technical staff. Thank you for your patience and apologies for any inconvenience caused.

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12623001050640p
Ethics application status
Submitted, not yet approved
Date submitted
4/07/2023
Date registered
28/09/2023
Date last updated
28/09/2023
Date data sharing statement initially provided
28/09/2023
Type of registration
Prospectively registered

Titles & IDs
Public title
A novel method using textural analysis of thermal images to predict the healing status of diabetic-related foot ulcers
Scientific title
Using textural analysis of thermal imaging to predict healing status of diabetes related foot ulcers
Secondary ID [1] 309991 0
Nil
Universal Trial Number (UTN)
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
diabetic related foot ulcer 330486 0
Condition category
Condition code
Metabolic and Endocrine 327334 327334 0 0
Diabetes

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
This research plans to codesign and validate a fast, computer-aided, non-invasive test using thermal imaging that can accurately predict healing of diabetes related neuropathic ulcers at week 12 from an image taken at week 0 (baseline).
Phase 1. Co-design process; there are 2 groups of participants:
a) Individuals living with diabetes related foot ulcers
b) Clinicians managing people living with diabetes who have a foot ulcer
The 2 groups of participants meet on separate occasions. The co-design process involves small focus group discussions guided by an experienced facilitator with a pre-determined schedule designed to explore experiences and opinions on the topic and make use of interactional group exchange to gather data less accessible through individual interviews. We aim to use this approach to explore participants thoughts and address their needs regarding using technology to collect data to enable the prediction of DFU healing. We anticipate that each focus group meeting will be about 1.5 hours. The approximate timeline between each session will be between 4 to 6 weeks.
The group discussions will run as follows:
Session 1: Will include clinician participants and start with a discussion of the ‘why’ this device is being developed by the research team. The following will be then be raised for discussion by the participants – although the content will be shaped by the input provided by the participants, as they may raise aspects not considered by the research team:
1. What would the key drivers be for clinicians to want to use such a device?;
2. Specifics around the device itself:
a. Range of costs they’d be prepared to pay for the device;
b. What weight of the device they would be willing to accept;
3. Specifics around using the device will be drawn from using a persona – a person with a wound who is coming to the clinic for their first appointment, and an assessment will need to be undertaken. Clinician participants will work through this first appointment to unpack the following:
a. The software of the device:
i. Does it need to be connected to their Smartphone or organisation's database?;
b. Specifics around the input/presentation of the information:
i. What is the easiest way to enter the meta-data;
ii. How should the information be presented ie. Binary – will heal/won’t heal?
c. What are the critical aspects of the outcomes that will dictate your usage of the device?
i. What would need to be included to guide the patient care plan?
This information will be then taken by the researchers and the device will be shaped by this input.
Session 2: Will involve clinician participants trialing the hardware to test it. Again, using a persona, clinician participants will be asked to consider: Weight, display, taking an image, entering meta-data, having the results displayed, and what staff need to do with the results so that it can be used in clinical care plans.
If significant issues are identified, this may need to be repeated after adjustments have been made by the engineers in the research team.
Session 3: Will involve people living with diabetes related foot ulcer participants, to provide their insights on what would be needed to make the device appropriate from their perspectives. Data obtained in these sessions will be used to adjust the prototype.
This session will start with a discussion of the ‘why’ this device is being developed by the research team. The following will be then be raised for discussion by the participants, asking them to consider as a person who is undergoing an assessment by the clinicians at the clinic – although the content will be shaped by the content provided by the participants:
1. What are your thoughts about the device and what it is trying to achieve?
2. To test the usage of the device with participants – to invite feedback on the process, from their perspective; and
3. How should the information identified by the device be presented to them?
Session 4: Will include clinician participants and the final version of the device will be generated including what people living with diabetes related foot ulcers have contributed. To review and finalise device prototype for the validation stage.
We anticipate that the validation study will commence 4-6 weeks after the completion of the co-design (phase 1) study.
Phase 2 - Validation study
Consecutive individuals with diabetes, aged 18 and over, who attend the diabetic foot and podiatry inpatient and outpatient clinics at Austin Health, Melbourne, Australia will be invited to participate.
Information will be collected from all participants at baseline, 1, 2 and 12 week follow up visits, and is expected to take between 10-15 minutes for each visit. Data collected during each visit include:
* Thermal images of the foot ulcer;
* Colour (RGB, red, green and blue) images of the ulcers;
* Wound dimension (area) tracing and depth; and
* Foot ulcer infection status (as per WIFI classification: 0 = no symptoms or signs of infection; 1: mild (less than or equal to 0.2cm cellulitis); 2: moderate (>2cm cellulitis/purulence); or severe (systemic response/sepsis)
Thermal and RGB (Red Green Blue) colour images and a wound tracing will be taken of participant foot ulcers during routine dressing changes by one trained person to keep the data collection consistent.
In addition, the following information will be collected:
* Patient demographics: age, gender;
* Past and current medical history, especially regarding their diagnosis and treatment of diabetes (eg type and duration of diabetes);
* Comorbidities;
* Their biochemical profile as clinically prescribed;
* Current medication list;
* The duration and description of the foot ulcer using SINBAD wound classification system characteristics (ie. Site, Ischaemia/Peripheral Arterial Disease (PAD), Neuropathy, Bacterial infection, Area, Depth);
* Documented risk factors for delayed healing
Thermal images (and therefore, the thermal imaging device) will only be used when taking thermal images at baseline, and then weeks 1, 2 and 12 (follow up visits).
Intervention code [1] 326408 0
Diagnosis / Prognosis
Comparator / control treatment
The following measurements have been listed because they will be recorded as part of usual practice.
These include colour images (RGB) of the ulcer;
Wound dimension (area) tracing and depth; and Foot ulcer infection status as per WIFI classification - 0 = no symptoms or signs of infection; 1: mild (less than or equal to .2cm cellulitis); 2: moderate (>2cm cellulitis/purulence); or severe (systemic response/sepsis). These methods are currently used in clinical practice and are considered as best practice for the identification of healing status of the ulcer.
Control group
Active

Outcomes
Primary outcome [1] 335274 0
Accuracy of a machine learning (AI) model using thermal images to predict delayed healing at 12 weeks. The texture of the thermal image of the wound will be analysed. The ground truth is that the wound is considered healed if at week 12, the ulcer has completely closed while it is considered unhealed if the ulcer has not completely closed.
Timepoint [1] 335274 0
Baseline (enrolment of participant) data collection
1 week post base-line
2 weeks post base-line
12 weeks post base-line
Secondary outcome [1] 423691 0
Colour images (RGB) of the ulcer. The current common practice of assessing wound healing is by comparing the area of the wound in week 4 to week 1. In a wound that is healing properly, the wound area should be reduced by 50% . We will verify that the current strategy of assessing healing wounds via RGB images using the area of the wound at week 4 is half of the week 1 wound area.
Timepoint [1] 423691 0
Baseline (enrolment of participant) data collection
1 week post base-line
2 weeks post base-line
12 weeks post base-line
Secondary outcome [2] 423692 0
Digital planimetry or the clinical ruler will be used to measure the dimensions of the wound and to compare dimensions of the wound as measured by RGB images.
Timepoint [2] 423692 0
Baseline (enrolment of participant) data collection
1 week post base-line
2 weeks post base-line
12 weeks post base-line
Secondary outcome [3] 424145 0
Foot ulcer infection status as per WIFI classification - 0 = no symptoms or signs of infection; 1: mild (less than or equal to 0.2cm cellulitis); 2: moderate (greater than 2cm cellulitis/ purulence); or severe (systemic response/sepsis)
Timepoint [3] 424145 0
Baseline (enrolment of participant) data collection
1 week post base-line
2 weeks post base-line
12 weeks post base-line

Eligibility
Key inclusion criteria
Phase 1: Clinicians who are working with people who have diabetic foot ulcers;
Phases 1 and 2: Adults diagnosed with diabetes;
Presence of a neuropathic foot ulcer;
Phases 1 and 2: Ability to provide consent;
Phases 1 and 2: Are available to participate over the study period.
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
Phases 1 and 2: Unable to provide informed consent;
Phase 2: Pregnancy;
Phase 2: Presence of infection/Osteomyelitis;
Phase 2: Severe cases of peripheral arterial disease (PAD) (toe pressure below 60 mmHg)

Study design
Purpose of the study
Diagnosis
Allocation to intervention
Non-randomised trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Masking / blinding
Who is / are masked / blinded?



Intervention assignment
Other design features
Phase
Not Applicable
Type of endpoint/s
Statistical methods / analysis
Phase 1: Co-design - thematic analysis
Phase 2: Validation
a. Primary analysis will involve evaluating the concordance between the model and true wound status at 12 weeks using Cohen’s Kappa for statistical agreement and reported with a respective 95% confidence interval
b. Receiver operating Curve Analysis will be used to report sensitivity;
c. Diagnostic odds ratio will be used to test the true status of wound healing at 12 weeks.

Recruitment
Recruitment status
Not yet recruiting
Date of first participant enrolment
Anticipated
2/10/2023
Actual
Date of last participant enrolment
Anticipated
20/12/2024
Actual
Date of last data collection
Anticipated
28/03/2025
Actual
Sample size
Target
120
Accrual to date
Final
Recruitment in Australia
Recruitment state(s)
VIC
Recruitment hospital [1] 25067 0
Austin Health - Austin Hospital - Heidelberg
Recruitment hospital [2] 25510 0
Royal Melbourne Hospital - Royal Park campus - Parkville
Recruitment hospital [3] 25511 0
St Vincent's Hospital (Melbourne) Ltd - Fitzroy
Recruitment postcode(s) [1] 41325 0
3052 - Parkville
Recruitment postcode(s) [2] 41326 0
3065 - Fitzroy
Recruitment postcode(s) [3] 40730 0
3084 - Heidelberg

Funding & Sponsors
Funding source category [1] 314170 0
Government body
Name [1] 314170 0
Department of health and Aged Care, Medical Research Future Fund
Country [1] 314170 0
Australia
Primary sponsor type
University
Name
University of Melbourne
Address
Swanston St PARKVILLE VIC 3052
Country
Australia
Secondary sponsor category [1] 316086 0
University
Name [1] 316086 0
RMIT University
Address [1] 316086 0
GPO Box 2476, Melbourne VIC 3001
Country [1] 316086 0
Australia

Ethics approval
Ethics application status
Submitted, not yet approved
Ethics committee name [1] 313298 0
Austin Health
Ethics committee address [1] 313298 0
Level 8, Harold Stokes Building Austin Hospital145 Studley RdHeidelberg , 3084 Victoria
Ethics committee country [1] 313298 0
Australia
Date submitted for ethics approval [1] 313298 0
01/05/2023
Approval date [1] 313298 0
Ethics approval number [1] 313298 0

Summary
Brief summary
Diabetes-Related Foot Ulcers are common in people living with diabetes and are a major cause of poor quality of life and disability. If they are not treated appropriately they may require hospitalisation and could lead to amputation and early death. Right now, we don’t have a quick and simple way to know which ulcers will heal normally, and which ones may delay healing, or lead to amputation. This project will trial a new way to analyse photography known as thermal imaging to see if there is extra information in this sort of image to help predict which ulcers will heal and which ones may not, so we can know which ulcers may need more aggressive intervention so they are more likely to heal.
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 127654 0
Prof Dinesh Kumar
Address 127654 0
RMIT UniversityGPO Box 2476, Melbourne VIC 3001
Country 127654 0
Australia
Phone 127654 0
+61 399251954
Fax 127654 0
Email 127654 0
dinesh.kumar@rmit.edu.au
Contact person for public queries
Name 127655 0
Prof Dinesh Kumar
Address 127655 0
RMIT UniversityGPO Box 2476, Melbourne VIC 3001
Country 127655 0
Australia
Phone 127655 0
+61 399251954
Fax 127655 0
Email 127655 0
dinesh.kumar@rmit.edu.au
Contact person for scientific queries
Name 127656 0
Prof Dinesh Kumar
Address 127656 0
RMIT UniversityGPO Box 2476, Melbourne VIC 3001
Country 127656 0
Australia
Phone 127656 0
+61 399251954
Fax 127656 0
Email 127656 0
dinesh.kumar@rmit.edu.au

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment


What supporting documents are/will be available?

Doc. No.TypeCitationLinkEmailOther DetailsAttachment
19607Informed consent form  dinesh.kumar@rmit.edu.au



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.