Early detection of melanoma – what are the challenges?

Melanoma (skin cancer) is one of the most common and aggressive cancers. Fortunately, melanoma is curable cancer that can be treated effectively, if detected early. Rapid diagnosis is essential to ensure that treatment is undertaken before local and advanced spreading occurs.

However, while surgical removal of melanoma is quite simple, early-stage melanoma detection is extremely difficult. Current practices require total body skin examination (dermoscopy) by a dermatologist, which is a time-consuming process, particularly as melanoma often resembles a common skin lesion (“mole”). Thus, the risk of missing a melanoma remains significant.

In the last  3 years, Artificial Intelligence (AI) systems for the identification of melanomas have become increasingly prevalent. However, most AI systems have yet to differentiate between normal tissue and abnormal skin lesions using images alone. This means that without integrated clinical data, systems continue to underperform compared to the average dermatologist.

Developing AI-based solutions for improving dermatologists’ diagnostic accuracy

Responding to this challenge, iToBoS aims to develop a patient-centred AI diagnostic platform for the early detection of melanoma.

The aim of the iToBoS project is to train an AI system with integrated information from various sources, ranging from dermoscopic images, medical records, family history and genomics data, to increase the precision of clinical decisions in the diagnosis of skin cancer.

iToBoS will develop and validate:

  • A novel total-body high-resolution scanner with results comparable to dermoscopy on a faster, automatic operation
  • A holistic AI cognitive assistant tool to empower healthcare practitioners to make a comprehensive patient-tailored diagnosis of skin cancer, leading to improved detection rates.

iToBoS will enable the classification of pigmented skin lesions to provide a reliable diagnosis and avoid burdening physicians with investigating non-cancerous skin growths.

Ethical, Privacy and Social Impact Assessment to Ensure Sustainable Innovation

Trilateral conducts a comprehensive privacy impact assessment of the iToBoS technologies to ensure the secure handling of patient data complying with existing regulations at both European and national levels.

Trilateral’s work also aims to maximise benefits and minimise the risks of AI technology with consideration to ethical and social values. In collaboration with all involved stakeholders, social and ethical challenges will be examined and used to improve clinician and patient understanding of the role AI can play in health care and to assist in the adoption of innovative technologies in the dermatological context.

Trilateral’s effort aims at increasing the sustainability of the project’s results by producing guidelines for healthcare professionals, patients, care- givers and -receivers and related EU projects.

Ensuring data ethics and algorithmic transparency

Trilateral will develop novel techniques for achieving algorithmic transparency and deal with issues such as algorithmic bias and interpretability. Trilateral will conduct an analysis of the AI models’ performance, pairing data ethics with AI evaluation assessments of the iToBoS tools. This will include an examination of the algorithms, system integration, and data management systems and their outputs.

For more information and updates follow us on Twitter.

EU flag yellow low e1523448262817
This project has received funding from the European Union’s Horizon 2020 Research and Innovation programme under grant agreement No 965221.

Learn more about our research in the field of


Let's discuss your career