Trilateral Research Tackles Bias in AI Skin Cancer Detection

Trilateral Research has partnered with healthcare providers, AI developers, and research institutions as part of the iToBoS Project (Intelligent Total Body Scanner for Early Detection of Melanoma), to create an innovative AI-powered screening tool for early melanoma detection. The system integrates AI with advanced imaging to enhance early melanoma detection and provide a more comprehensive view of skin health.

A key challenge in AI-driven skin cancer detection is bias, as many existing tools rely on datasets that lack diversity. This limitation increases the risk of inaccurate or delayed diagnoses, particularly for underrepresented groups. Trilateral is focused on minimising bias in the iToBoS tool, making it more equitable, transparent, and effective across diverse skin types. The team is also working with partners to reduce bias in AI-driven skin cancer detection, to ensure consistent and equitable outcomes for all patients.

Trilateral evaluated the AI-driven system within the iToBoS project to ensure it operates fairly, transparently, and inclusively, with a diverse and representative dataset. A crucial part of this effort is the Risk Assessment Model, developed by the National Technical University of Athens (NTUA), which measures the reliability of the AI’s diagnostic decisions and identifies areas for improvement. Trilateral’s experts provided targeted recommendations to improve inclusivity, strengthen decision-making, and ensure the tool delivers accurate and fair diagnoses for all patients.

Trilateral and its partners are refining how the AI processes and learns from patient data, expanding its ability to accurately detect melanoma across all skin types. By incorporating best practices for fairness and transparency, they are ensuring the system is both reliable and adaptable to real-world clinical settings. The iToBoS project aligns with Trilateral’s broader mission to ensure AI technology is used responsibly and meets ethical and regulatory standards.

The iToBoS project is shaping the future of medical AI. By setting new standards for fairness and responsible deployment, it is paving the way for more inclusive and effective AI-driven healthcare solutions. Trilateral Research remains dedicated to advancing AI solutions that prioritise patient safety, accuracy, and fairness, ensuring that emerging technologies benefit all communities equitably.

If you’d like to find out more about Trilateral Research, it’s products, services or research innovations, get in touch.