AI-MEL: Image Analysis and Machine Learning for Early Diagnosis and Risk Prediction in Children, Adolescents and Young Adults

What this Trial Is About

Researchers are developing supportive artificial intelligence (AI) algorithms to help distinguish melanoma from nevi or other benign pigmented skin lesions, focusing especially on younger patients under 30 years old. The study aims to improve early diagnosis and risk prediction of melanoma, targeting children, adolescents, and young adults. The project is led by the German Cancer Research Center and focuses on vulnerable populations in skin cancer screening. The study involves creating two AI algorithms: one based on dermatoscopic images for skin cancer screening and another based on histological images to assist pathologists when lesions are still suspicious after dermatologic evaluation. The research also includes developing explainability methods to help users understand the AI decisions better, reduce biases, and increase trust in these tools. There is no additional treatment or intervention for participants, as all data come from routine clinical practice. Participants' involvement requires no extra time since the study uses data collected during standard care. Researchers will evaluate the AI models by measuring the Area Under the Receiver Operator Curve (AUROC) and balanced accuracy at the end of the first training and testing cycle (about 1.5 years from the study start), with reevaluations at 6 and 12 months for improvement. This observational study continues until November 2026 and includes healthy volunteers of all ages and genders.

Artificial Intelligence Based Melanoma Early Diagnosis and Risk Prediction in Children, Adolescents and Young Adults