Present day chronic wound measurement is archaic and outdated. The most common approach for tracking wound progression includes a medical aid or nurse estimating maximum width, height, and depth with a paper tape measure. More sophisticated approaches use overhead transparencies and markers to manually trace a given wound profile.
Both procedures are primitive, and in some ways, can even be considered harmful. Estimated nominal wound dimensions in tandem with flattened, two-dimensional measurement provides a curve that is not characteristic of real-life healing. Chronic, deep tissue wounds heal from the bottom up, for example, meaning properties may change significantly before a 2D projected profile would note any meaningful delta. Furthermore, time spent performing manual measurements removes caretaker’s attention from more pressing, less automatable, tasks.
Diabetic foot ulcers are particularly severe as symptomatic patients are likely to experience discomfort, frequent clinical visits, and in the worst case: amputation. A thorough understanding of a given treatment with respect to a patient’s specific physiology, as well as a trendline describing heal rate over time, is fundamental for making informed clinical decisions. In the present day, no products exist on the market to seamlessly bridge the gap between at-home patient experiences and clinical expertise.
Wound³ is an iOS-based imaging software capable of measuring and tracking wound progress over time. In less than one second, or simply the time it takes to snap a photo, the platform calculates and standardizes wound properties – without any cloud computation or physical markers for camera calibration. In one sentence: Wound³ transforms any iPhone into a standalone high-precision medical scanner.
After a measurement is captured, AI-endowed analysis calculates contoured surface area, volume, and depth, while rendering the wound in 3D space. Colour data is also captured and appended to the 3D model and scan metadata, assisting a medical professional’s ability to predict hyper granulation, infection, and necrosis. Qualitative data is also collected through Wound³’s ‘Journal Entry’ functionality, which provides a portal for patients to dictate symptoms, as well as answer standardized questions related to tenderness, smell, drainage, inflammation, and more.
All data (both qualitative and quantitative) can be exported to PDF format and provided to one’s healthcare provider or EMR. This PDF includes graphs, images, 3D renders, and Journal Entries, all of which paint a holistic picture of the wound’s past and present, to accurately predict the future.
At the present date, Wound³ has a functional beta application. With this said, our AI systems are only compatible with silicone wound models. To generalize the AIs, real patient data must be collected through a large clinical study. The purpose of the study would be to simply collect thousands of anonymized data samples depicting diabetic wounds in various stages and environments. Furthermore, racial diversity and wound position must also be considered for an unbiased system.
It is important to note that the clinical study, and thus the first iteration of the platform, will be specific to diabetic ulcers. The reason is twofold: diabetic ulcers are common and severe (which places this complication in the upper echelons of a triage-style importance hierarchy), and ulcer boundaries are more approachable from a technical perspective (as boundaries are more well- defined relative to complications such as burns, which can manifest as gradients).
After the clinical data is collected, the success of the AI’s inference capacity will be tested by comparing AI-predicted wound boundaries with human-segmented alternatives. A high correlation between automated and professional results suggests a reliable system.
For more information regarding the design of our study, please reach out to us directly through the contacts portal on this website.
Besides the obvious diagnostic aid our platform provides, from a general perspective, portable, scalable medical imaging has the ability to revolutionize patient-practitioner interaction during treatment.
Wound³ blatantly rejects of all things proprietary – and for good reason. Statistically speaking, 60% of Canadians can leverage Wound³’s technology – with the devices in their pockets – today. Most other companies competing in the chronic wound measurement space focus on hospital-specific instruments. Technologies such as infrared cameras, photogrammetry, and cloud computing are leveraged to model the wound computationally – which is slow, expensive, and most importantly: not available outside of the confines of a multi-billion-dollar institution.
High measurement frequency catalyzes excellent decision making. The only way to scale measurement frequency is through decentralization; medical devices must be ubiquitously available in a non-institutional setting. Wound³ provides this service, placing patients, at-home caretakers, and clinicians equally at the forefront of analysis in a streamlined manner.
Lastly, Wound³ comfortably integrates into the COVID inspired world of telemedicine. All collected data can be easily visualized by a certified professional in a virtual environment, allowing the patient to communicate with the care provider digitally.
Wound³’s scanning architecture is patent-pending and non-affiliated with any organizations outside of the company, including the University of Alberta.
Go to Market Plan
A dual-front rollout plan will be employed to maximize potential user retention and long-term revenue.
From a public health care perspective: Wound³ will seek Epic certifications (FHIR, HL7), then integrate into Connect Care for AHS EMR integration. This is a long-term plan and will take years to accomplish.
From a private health care perspective: geriatric centres and private clinics will be contacted, with the primary value proposition being time saved on a per-measurement basis. When the clinical study commences, Wound³’s marketing efforts will focus on generating Letters of Intent contingent on the measured AI inference accuracy.
For a detailed analysis describing either action plan, please reach out through the contacts portal below.