Training for change, from science to practice.

Sprint AI Training for African Medical Imaging Knowledge Translation (SPARK) is committed to empowering African data science communities and imaging researchers to develop deployable imaging diagnostic tools that can assist clinicians to provide high-value care and bridge the gap in the medical disparity between Africa and the rest of the world.

SPARK is one of the training platforms of the Consortium for Advancement of MRI Education and Research in Africa (CAMERA) offered through the Medical Artificial Intelligence (MAI) Lab at Crestview Radiology Ltd, Lagos, Nigeria and the Multimodal Imaging of Neurodegenerative Disease (MiND) Lab, McGill University, Montreal, Canada.

The SPARK program, endorsed by the Medical Image Computing and Computer Assisted Intervention (MICCAI) Society, represents a comprehensive training program aimed at building a community of experts in artificial intelligence (AI) for medical imaging across Africa.

To achieve our mission, we provide a free hands-on training program that implements the RAD-AID, Teach-Try-Use approach and takes trainees through three learning blocks to successfully train-the-trainers and create the critical mass of AI imaging experts in Africa.

SPARK uses a curriculum that is case-based and a train-the-trainer model to provide free access to hands-on skill development. This model adopts the RAD-AID Teach-Try-Use strategy, comprising foundational courses (Teach), a hybrid hackathon (Try), and a guided practicum (Use). This provides participants with theoretical knowledge, hands-on experience, and practical application skills.

Each SPARK Academy is aimed at a real-world use case and implements collective intelligence built around a networking space to advance AI capacity and rapid implementation in Africa.

Our first SPARK Academy, endorsed by the Medical Image Computing and Computer Assisted Intervention (MICCAI) Society, trained African clinicians and scientists to develop state-of-the-art AI methods for glioma brain tumor segmentation.

The first program brought together oncology, neurology and neurosurgery, radiology, medical physics, biomedical engineering and computer science residents and graduate students from across Africa who co-developed informed AI methods for segmentation of brain tumors into sub-regions on magnetic resonance imaging (MRI) scans.