Chronic wounds present a major public health challenge with significant health and economic costs [1]. Chronic wounds are those that have not progressed through the normal stages of healing in a timely and organized process [2] and substantially reduce the quality of life of those affected [3]. The most common types are pressure injuries and chronic lower leg wounds, including diabetic, venous, and arterial ulcers [4]. The global burden of skin and subcutaneous diseases has risen rapidly over the last 15 years, but the lack of international prevalence data for chronic wounds limits the ability to estimate true impact [5]. Based on available figures, an estimated 2.5% of Americans are affected by chronic wounds, the majority of whom are older adults [6]. In Australia, the prevalence of chronic wounds is estimated at 1.9% [7] and accounts for Aus $6.6 billion (US $4.29 billion) per year in costs to health and aged care budgets [8]. Between 2013 and 2017, more than two-thirds of people with chronic wounds who were admitted multiple times to public hospitals in the Australian state of New South Wales (NSW) were aged 65 years or older [9].

Older people, with increasing frailty, impaired skin integrity, and comorbidities such as diabetes, obesity, and peripheral vascular disease, are disproportionately affected by chronic wounds [10,11]. Chronic wounds are particularly common among residents of nursing homes, with an estimated prevalence of pressure injuries of up to 26% [12]. Despite being common, less than 1 in 5 of those with pressure injuries had a wound management record documenting every current injury in a 2018 NSW study of 67 nursing homes [12]. Inadequate wound prevention and treatment was also emphasized by the 2021 Australian Royal Commission into Aged Care Quality and Safety as causing pain, distress, and premature death [13]. Aged care staff giving evidence at the commission reported inadequate knowledge and training in managing pressure injuries effectively [13]. Despite the commission’s recommendation that wound care training and education be improved, inadequate wound care continues to be one of the most common clinical complaints made to the Australian Aged Care Quality and Safety Commission [14].

Evidence-based wound care can improve clinical outcomes of chronic wounds [15]; however, provision of such care in nursing homes can be challenging. Wound care is a specialized field, and in aged care, there is higher variance in wound management and less specialist wound oversight than in other settings [10]. Chronic workforce shortages, high staff turnover, variable wound care expertise among staff, and fragmented and untimely access to expert wound care also limit the provision of consistent and high-quality wound care [13]. Along with evidence-based care, access to clinicians with expertise in wound management can ensure high-quality care and improve resident outcomes; however, in Australia, specialized wound care clinics are generally confined to major centers, which limits in-person attendance by rural and immobile nursing home residents [16]. This geographic centralization of wound care services creates significant barriers to timely assessment and treatment, contributing to delayed wound healing and avoidable complications [17].

Virtual care, health care delivered remotely using information and communication technology services [18], can address access barriers to ensure personalized and timely health care [19] for those unable to attend in-person wound clinics, such as nursing home residents. While the terms virtual care and telehealth are frequently used interchangeably, virtual care is a broader term encompassing telehealth (video or telephone consultations), remote patient monitoring, store-and-forward communications, and website and mobile apps [20]. Virtual care has demonstrated safety and effectiveness in wound healing and can overcome access gaps in chronic wound care [21,22]. Virtual wound care trials have also shown reductions in wound healing times and wound-related hospitalizations [23]; however, there have been few studies undertaken in nursing homes [24]. Furthermore, while current virtual wound care interventions offer wound analysis and measurement capabilities, they lack seamless integration with video telehealth platforms, restricting their effectiveness in remote care.

The number of Australians living in permanent aged care is growing, with 178,000 people aged 65 years or older residing in aged care in 2022, a 3.1% increase since 2017 [25]. Australia has among the highest proportions of older people living in permanent aged care when compared with other similar countries, including the Netherlands, Japan, and Canada [26]. Despite increasing numbers of nursing home residents, there has been a 5.2% drop in aged care providers, attributed to regulatory and financial pressures [27]. To meet the industry’s growing demands and address the imminent implementation of the strengthened Aged Care Quality Standards [28], virtual care innovation, such as artificial intelligence (AI) to assist with wound analysis and remote consultation, is needed to transform the aged care sector, meet the needs of consumers, and improve care outcomes [27]. One such virtual care solution is WoundView, a new AI wound analysis application embedded in an existing telehealth platform, Coviu, which aims to address access barriers to evidence-based and specialized wound care.

A 5-year, multipartner research project funded by the Australian Government’s Medical Research Future Fund has developed the WoundView application (here referred to as WoundView) to improve access to specialized wound care in nursing homes and other remote settings. WoundView uses a computer vision–based AI algorithm that has been developed and validated using more than a thousand wound images. WoundView produces wound bed segmentation and measurements (ie, length, width, and total area), which can be monitored and compared over time. WoundView (Figure 1) is embedded in the existing telehealth platform, Coviu, allowing for real-time video telehealth between nursing home residents and staff and external wound clinicians. The Coviu platform enables the video call component of the Australian Government–funded Healthdirect Australia, the national virtual public health information service, enabling telehealth use in Australian nursing homes at no cost.

WoundView uses consumer-grade, hand-held devices, including mobile phones, laptops, and tablet devices to assess chronic wounds. The Coviu platform delivers secure, web-based video telehealth. It does not require the download of an app, and all transmitted call data, including images and documents, are transmitted peer-to-peer (ie, between call participants) and protected with end-to-end encryption [31]. This protocol describes a pilot and feasibility study to assess the acceptability and feasibility of WoundView for residents and staff working in nursing homes and external wound care clinicians in NSW, Australia.

The primary objective of the pilot study is to assess the feasibility and acceptability of the proposed approach in preparation for an implementation study on WoundView. Study recruitment and retention and protocol adherence and adaptations will be used to evaluate the primary objective. The secondary objectives are to assess the acceptability of data collection tools for possible inclusion in the implementation study and evaluate collected wound images to verify and update the WoundView AI wound analysis algorithms for future use in nursing homes.

The WoundView pilot and feasibility study is a prospective, nonrandomized trial in which all residents with a chronic wound will be eligible to receive the intervention, WoundView. Use of WoundView will be facilitated by staff working at pilot site nursing homes, and wound care will be provided by the Wound Care Command Centre (WCCC) or usual wound care providers. Nursing home staff, WCCC clinicians, and usual wound care providers who use the intervention will also be invited to participate in the study.

The study will determine the acceptability of WoundView and assess the feasibility of selected outcome measures in preparation for an implementation study. The pilot and feasibility study will be reported using the CONSORT (Consolidated Standards of Reporting Trials) extension to pilot and feasibility trials Checklist 1 [32], SPIRIT (Standard Protocol Items: Recommendations for Interventional Trials) statement Checklist 2 and additional guidance on reporting of nonrandomized pilot and feasibility studies by Lancaster and Thabane [33]. A mixed methods approach will be used to assess acceptability and feasibility.

The study will involve 2 phases (Figure 2). During Phase 1, staff working in the pilot site nursing homes will be upskilled in general telehealth and trained to use WoundView. Phase 2 will pilot WoundView and assess the feasibility of the trial design for the planned implementation study.

The study will be conducted in 2 private, not-for-profit nursing homes in NSW, Australia. One site is an 80-bed facility in outer regional NSW, as defined by Australian Statistical Geographical Standard Edition 3 [34]. The second site is a 140-bed facility in Sydney, NSW, a major city [34]. The pilot study nursing home sites will be supported by either the WCCC, a team of specialized nurses at the Royal Prince Alfred Virtual Hospital based in Sydney, Australia, or their usual wound care provider.

The study will recruit three participant groups: (1) nursing home residents living with chronic wounds, (2) nursing home clinical staff involved in wound care (eg, registered nurses, enrolled nurses, and wound care coordinators), and (3) WCCC clinicians and usual wound care providers, (eg, wound care nurses and general practitioners) who are providing specialized wound care using WoundView.

The intervention, WoundView, is a novel computer vision–based AI wound analysis application integrated into the Coviu telehealth platform. WoundView is a software component that extends the functionality of the Coviu platform, without requiring changes to its core code. WoundView will link residents with chronic wounds and their clinical care team to specialized wound care clinicians at the WCCC in Sydney, Australia, or their usual wound care providers, who will deliver expert wound care based on clinical data gathered during video telehealth, AI wound analysis, including wound measurements and segmentation, and resident and family preferences.

WoundView requires an initial referral from the participating nursing home to the WCCC or usual care provider for an eligible resident. An appointment is scheduled using the Coviu platform, and the resident, nursing home staff, and wound care clinician attend at the scheduled time. At each use of WoundView, wound image and video data will be gathered by nursing home staff who will take a series of wound photographs prior to or during the WoundView call and submit these for AI wound analysis. The inclusion of an ArUco marker (a marker of known size) in each wound image enables AI wound size analysis using the WoundView AI algorithm. The AI wound analysis provides wound measures such as length, width, and area, along with wound segmentation. Live video stream of the wound will also be available during the call, providing treating clinicians with detailed, close-up vision of the wound, an innovative feature of WoundView. The wound care clinician provides clinical advice about wound management, including cleaning, debridement, dressings, and frequency of dressing change. Further wound review is arranged as advised by the wound care clinician.

The primary outcome will be a narrative description of study feasibility, including participant recruitment and retention, protocol adherence and adaptations, and stakeholder acceptability (Table 1). A traffic light system will guide progression to an implementation study [37], red (stop, main study not feasible), amber (feasible with modifications or close monitoring), and green (continue without modifications) [38]. A recruitment rate of greater than 50% (defined as the percentage of eligible residents recruited) and retention of greater than 75% (proportion of participants who remain in the pilot for its duration) will be considered acceptable to proceed with the implementation study without modification. Protocol adherence, the degree to which the WoundView intervention and research protocol were implemented as intended, greater than 75% will be considered acceptable. Additionally, regular stakeholder input and agreement on the interpretation of pilot findings will be sought [37]. Acceptability will be assessed by measuring participants’ satisfaction with study assessment instruments, and the use, delivery, barriers, and enablers to the intervention, WoundView. Data collection instruments are described in the section below. Secondary outcomes, including wound measurements, virtual care activity, health-related quality of life (HRQoL), hospitalization rates, health resource use case studies, and patient and staff satisfaction, will be assessed for feasibility of inclusion in a planned implementation study.

ED visits and hospital admission data, including length of stay, reason for admission, diagnosis on discharge, and discharge location, will be recorded and analyzed using descriptive statistics. Descriptive statistics will also be used to illustrate the change in wound size across the intervention period as well as to summarize HRQoL, AEs, and satisfaction survey results. All analyses will be intention-to-treat. An estimate of costs will be derived from the total number of dressing changes and extrapolated as per Wilson et al [55]. The actual and projected treatment costs will be calculated and extrapolated to estimate the overall cost of wound care in the pilot nursing home sites. Estimated costs will be compared with other Australian estimates in nursing home settings [56].

Wound images collected during the WoundView pilot and feasibility study will be used as part of the verification and update of the WoundView AI wound analysis algorithm. To verify the quality of the AI wound analysis algorithm, a ground truth mask and an AI-predicted mask will be created and compared for each wound image. Masks are pictorial representations that isolate a specific part of an image, such as the wound bed from the background. To create the ground truth mask used to verify the performance of the AI wound analysis algorithm, wound images collected using WoundView during the pilot will be manually annotated by experienced wound care clinicians using the Computer Vision Annotation Tool (Intel Corporation), a free, open-source, web-based image and video annotation tool. Random checks of the annotation accuracy will be conducted by 3 wound care experts. To verify the performance of the AI model, ground truth annotations will be compared with the AI-predicted mask. Multimedia Appendix 1 outlines data descriptions, study devices, data handling, and storage processes, as well as data retention for the study. Explicit consent will be sought from participating residents for the use of captured wound images in the verification and update of the WoundView AI wound analysis algorithm.

Conduct of the pilot will be overseen by a data monitoring group consisting of members from the research team from the university and Coviu. The Chief Investigator is a member of the data monitoring group and will make the final decision on ceasing or proceeding with the trial. The group will meet regularly to discuss pilot data quality and completeness. The group will review source documents (eg, wound assessments, virtual health activity, health resource use case studies, and HRQoL measures) along with compliance with informed consent and eligibility criteria and monitoring of AEs and endpoints. Additionally, data storage and handling practices, particularly with respect to the transfer of wound images (Multimedia Appendix 1), will be regularly reviewed. If a problem is identified (eg, missing study documents), the team will work to resolve the issue. All parties involved will keep participants’ data strictly confidential.

The trial has been registered with the Australian and New Zealand Clinical Trials Registry on June 2, 2025 (ACTRN12625000565448). Ethical approval was received from the University of Sydney Human Research Ethics Committee (HREC; 2024/HE001356). Informed written consent will be obtained from all participants. Participants will not receive any payment or reimbursement for their participation. Participants may withdraw from the research study at any time with assurance that their withdrawal will not affect their relationship with the nursing home, employer, or the university. The pilot and feasibility study will collect personal information, which will be reidentifiable and securely stored in HIPAA (Health Insurance Portability and Accountability Act)–compliant REDCap (Research Electronic Data Capture; Vanderbilt University), a secure web application for managing online surveys and databases hosted by the University of Sydney in NSW, Australia. Only approved members of the research team will have access to personal information. All information will be subject to the University’s HREC confidentiality policies. Figure 1 uses 2 images of people. These are generic images and are not of research participants. The image of the clinician using the laptop was obtained with permission from Coviu Global Pty Ltd (developer of WoundView) [29]. The image of the older woman and nurse using a mobile device was sourced from Unsplash [30], which grants “irrevocable, nonexclusive, worldwide copyright license to download, copy, modify, distribute, perform, and use images from Unsplash for free, including for commercial purposes, without permission from or attributing the photographer or Unsplash.”

Any required changes to the protocol will be submitted as amendments to the University of Sydney’s HREC for approval before implementation. In an emergency, deviations from the protocol may be made to protect human subjects’ rights, safety, or well-being without the HREC’s prior approval. Such deviations will be promptly documented and reported to the HREC.

Results of the study will be included in reports to stakeholders, journal publications, and conference presentations as part of the dissemination of the broader study’s outcomes. Findings will be provided in such a way that participants cannot be identified.

Chronic wounds are common among nursing home residents, detrimentally affecting their quality of life and contributing to increased risk of premature death. When compared with other populations, nursing home residents experience significant barriers accessing timely and high-quality wound care. While research in virtual wound care interventions is rapidly expanding, little is known about its use in nursing homes despite the high burden of disease and substantial impact on health outcomes.

The development of a novel computer vision–based AI wound analysis application embedded in an existing telehealth platform is warranted in offering clinicians and residents a “one-stop shop.” It is expected that the findings of this study will provide sufficient methodological evidence regarding the design and planning of the intervention to warrant an implementation study. The methodological evidence will include an approach to participant recruitment, retention, and adherence to the intervention. Although this pilot and feasibility study is not powered to detect statistically significant effects, the findings will provide insights to optimize the design of a larger scale implementation study and strengthen the research approach. Results will also inform the implementation and evaluation of new virtual wound care interventions in nursing homes.

The pilot study will assess the acceptability and feasibility of a novel wound care telehealth intervention, WoundView, and will contribute to the design of an implementation study testing WoundView in a range of nursing homes in Australia. The integration of WoundView is expected to transform the use of telehealth for wound management and lead to earlier intervention and better access to specialist wound care services for nursing home residents.