Global Medical AI Competition with a USD 9,000 First Prize

The Radiological Society of North America (RSNA) in collaboration with Kaggle offers the RSNA Knee Abnormality Detection 2026 competition. It aims to develop machine learning models capable of detecting a range of knee-related abnormalities in MRI scans. Participants build multimodal models using medical images and radiology reports to detect 12 types of conditions, including ligament tears, meniscal injuries, osteoarthritis, fractures, inflammation, and joint effusion.

Prize money for top placements; potential recognition within the data science and medical imaging communities.

Open to individuals and teams worldwide; a Kaggle account is required to participate.

Submit entries through Kaggle notebooks following the official competition rules and data licensing.