Including Indigenous data to train AI models on dermatological diseases for early detection

Use of Canadian data, including Indigenous data, to train AI models on skin cancer and other dermatological diseases for early detection in primary care settings.

Building a secure Canadian skin-health dataset so AI can help doctors spot high-risk cases sooner.

Canada faces one of the world’s highest per-capita rates of skin cancer, yet access to dermatology care is limited especially in rural, northern, and Indigenous communities. Primary-care teams often must make referral decisions without timely specialist input, which can delay diagnosis and increase strain on the health system.

This project will establish a Canada-wide, privacy-preserving skin-health data network to support earlier detection and timely patient care, that way patients can get the treatment they need, when they need it. Participating sites will bring together de-identified clinical information, pathology results, and medical images, and may include key social and environmental context to better understand risk and outcomes across diverse populations.

Privacy and trust are foundational. Data will be de-identified or pseudonymized, encrypted, and managed in accordance with provincial privacy requirements and Indigenous data-sovereignty principles.

Using data models to standardize information across jurisdictions, we will apply AI and machine learning to refine and evaluate dermatology decision-support models trained on Canadian data. The goal is to help primary-care teams assess skin condition risk and referral urgency more consistently, while testing performance across regions, devices, and populations.

By the end of the project, we will deliver a working national network, validated models with fairness and robustness checks, and a clear path to piloting in real-world clinics. The expected impact is faster, more equitable triage, fewer delayed diagnoses, and stronger Canadian leadership in responsible health AI.

Our mission has always been to increase accessibility to healthcare for all. For AI in healthcare, trust comes from evidence. We’re not just building models—we’re validating them across diverse populations and real-world settings so the benefits reach everyone.” - Maryam Sadeghi, CEO, MetaOptima