23 November 2021
Australia has one of the highest rates of melanoma, and melanoma is one of the deadliest forms of skin cancer. The earlier you find it, the greater the chances of successful treatment and the lower the cost. That is why Queensland researchers are pioneering 3D total body imaging for melanoma detection, disrupting traditional clinical models of care.
Leading the charge is world-renowned dermatologist Professor H. Peter Soyer. Professor Soyer identified the potential for 3D scanning technology previously used for cosmetic surgery, for melanoma detection.
Professor Soyer said melanoma can often be difficult to diagnose. Combined with a limited number of dermatologists to cover Australia’s geography, the burden on doctors to identify this type of skin cancer is large. Ultimately, we aim to provide more equitable access to advanced skin cancer screening, especially for people living in remote and regional communities.
“Our research is part of a broader program of work happening here in Queensland and with interstate colleagues at the University of Sydney and Monash University through the Australian Cancer Research Foundation (ACRF) and Australian Centre of Excellence in Melanoma Imaging and Diagnosis (ACEMID) cohort.
“ACEMID is establishing 15 research nodes using 3D scanners across Australia. Our clinics in Queensland are contributing to data assets that we can use to develop next-generation skin cancer diagnosis,” Professor Soyer said.
Rapid Applied Research Translation (RART) funding helped the Queensland research team establish its first clinic at the Princess Alexandra Hospital and focus on informatics. Professor Soyer considers RART instrumental in bringing dermatological informatics into the team’s original project plan.
“RART enabled us to compare diagnosis and clinical decision-making between teledermatology and face-to-face dermatology assessment, which will guide the future development of this system. This is funding well-invested when you consider how to expand this service across the country. In Brisbane, so many patients volunteered for 3D scanning that the clinical trial is fully booked until next year. For regional areas, the 3D scanner and teledermatology will be a game-changer. It will save time and travel expenses, reduce appointment times and healthcare costs and save lives.
“Consumer interest in this research has been high. Earlier surveys showed 99% of participants agreed 3D body imaging was beneficial in improving skin cancer diagnosis, and in the monitoring and teaching of skin conditions. This contributed towards a consumer-driven direction for future research.
“At the PA Hospital, most patients are seen by registrars under the supervision of consultant dermatologists. One of the project’s objectives is to train the next generation of dermatologists. Our ACEMID colleagues in New South Wales and Victoria have also installed the technology, but, unfortunately, due to COVID-19, they have not been able to screen many patients yet.
“We hope the results of our initial trials will help to inform other research projects around melanoma, such as the work of Associate Professor Victoria Mar, Professor Pablo Fernandez-Penas and Professor Monika Janda.
“There is a fair bit of momentum behind melanoma research in Queensland. Part of our project’s success is because of the diversity of the people involved. They bring in knowledge from a range of specialities,” Professor Soyer said.
Integration with Australia’s telemedicine network
Professor Soyer has teamed up with Associate Professor Liam Caffery, an expert in telehealth, to open up new possibilities to combine the innovative 3D scanning technology with the latest telehealth systems.
Associate Professor Liam Caffery explained how it works.
“The 3D scanner has a front and rear gantry and 92 cameras pointed at different angles at the patient’s body. The software renders the photos taken in milliseconds into a single avatar, which the clinician can view, rotate and zoom into individual moles on the skin.
“It is almost a full skin service. For privacy, the patients keep on their underwear, and the machine does not take images of the soles of the feet, so these areas cannot be screened. It also can’t screen a scalp covered with hair. But remember that melanoma usually occurs on parts of the body that have been overexposed to the sun.
“The 3D scanner allows us to create a more accurate picture of skin changes over time. This is very hard for doctors to do without taking photographs of every lesion. Some people have hundreds of moles on their body and doctors can’t keep track of every one of them.
“For regional areas, where access to skin cancer services is limited, we will use the 3D scanner to screen patients and beam their images to a remote dermatologist, most likely in Brisbane, for diagnosis. We hope that Mt Isa Hospital will house our first remote 3D scanner.
“After reviewing the avatar, the dermatologist will send the diagnosis back to the patient’s local doctor for treatment or surveillance of any suspicious lesions. The system will allow doctors to investigate and track potentially problematic skin spots over time.
“In the future, once we have available data sets, we can start using artificial intelligence, training algorithms to identify melanoma. The technology is not there yet, and there is work to do around creating appropriate standards. Through the International Skin Imaging Collaborative, we have access to an open-source archive of images of skin cancer. But this archive does not include the likes of the avatars we are developing.
“This system provides much richer data and allows us to identify other risk factors, such as hair and skin colour or degree of freckling. In existing datasets, there is a photographic bias. Typically, we only take photos of lesions that are suspicious. We need images of normal and abnormal lesions to train AI systems,” Associate Professor Caffery said.
From here, the research team plans to determine how 3D body scanning can integrate effectively and appropriately with Australia’s evolving digital health ecosystem.
Professor Soyer said there is still research to do before this technology is translated into clinical practice.
“In five to ten years, AI will have kicked in, and 3D imaging and teledermatology will have evolved. But there’s a lot to resolve before this happens. There are technical limitations to what we can achieve with 3D scanning. We need to do more research to resolve workflow inefficiencies, and finally, we need to recommend appropriate governance models, privacy frameworks and policies. None of this is straightforward, but it is exciting because of the system’s potential to increase access to specialist dermatology care, improving the early detection of melanoma and saving lives,” Professor Soyer said.
Project Investigators
Project leaders engage with and draw on the expertise of partners within universities, research institutes and hospital and health services around Queensland.
Publications:
Clare Amy Primiero, Aiden M McInerney-Leo, Brigid Betz-Stablein, David C Whiteman, Louisa Gordon, Liam Caffery, Joanne F Aitken, Elizabeth Eakin, Sonya Osborne, Len Gray, B Mark Smithers, Monika Janda, H Peter Soyer, Anna Finnane, Evaluation of the efficacy of 3D total-body photography with sequential digital dermoscopy in a high-risk melanoma cohort: protocol for a randomised controlled trial, BMJ Open. 2019 Nov 10;9(11): e032969. doi: 10.1136/bmjopen-2019-032969.
Jenna E Rayner, Antonia M Laino, Kaitlin L Nufer, Laura Adams, Anthony P Raphael, Scott W Menzies, H Peter Soyer, Clinical Perspective of 3D Total Body Photography for Early Detection and Screening of Melanoma, Front Med (Lausanne). 2018 May 23;5: 152. doi: 10.3389/fmed.2018.00152. eCollection 2018.
Useful links:
In the Media:
Quicklinks
Health Translation Queensland acknowledges the Traditional Owners and their custodianship of the lands on which we live, work, and play. We pay our respects to their Ancestors and their Descendants, who continue cultural and spiritual connections to Country. We recognise their valuable contributions to Australian and global society.