The Sprint AI Training for African Medical Imaging Knowledge Translation (SPARK) program is designed to train a new generation of African artificial intelligence (AI) experts in medical imaging who can train others. The training methodology uses a case-based learning approach and collective intelligence from various related fields (radiology, medical physics, computer science, neurology/neurosurgery, and oncology) to rapidly turn AI knowledge into action.

The first SPARK Academy held this year from April 3 to July 31, 2023 is in collaboration with McMedHacks and is endorsed by the Medical Image Computing and Computer Assisted Interventions (MICCAI) Society. This first SPARK Academy event is planned to coincide with the release of the AfNiA (The African Neuroimaging Archive) brain tumor dataset (BraTS-Africa) in May 2023, ahead of the 2023 MICCAI Brain Tumor Segmentation (BraTS) Challenge. The theme for this year’s program, dubbed BrainHack 2023, is “Train for Change, from Science to Practice”.

Ten (10) training sites in eight (8) African countries hosted trainees in the program, consisting of a 6-week virtual training (from April 3 to May 13, 2023), followed by a 1-week hybrid virtual/physical hackathon at the local sites. The third phase of the program will run from May 29 to July 31, 2023 during which period trainees are expected to apply the skills gained over the first 7 weeks to develop award winning models for brain magnetic resonance (MR) image segmentation.

After a competitive selection process from among a large pool of applicants, 5 trainees were recruited from the host country and 2 trainees from two other African countries that were not within the participating countries. The Diagnostic Imaging Research Centre (DIRC) of the University for Development Studies (UDS) hosted the Ghanaian Team on the Dungu-Tamale campus of the UDS for the weeklong hybrid virtual/physical hackathon from May 22 to May 26, 2023.

At the end of the project, participants will be trained to create state-of-the-art AI methods for brain tumor diagnostics. This first cohort will work together to produce AI methods for the 2023 MICCAI BraTS Challenge, with top ranked methods eligible for The Lacuna Fund monetary prize (up to $5000).

The main objectives of the SPARK Academy are to:

1. build an interdisciplinary capacity of highly qualified AI personnel across Africa who can design, implement, and evaluate AI methods for medical imaging.
2. provide transferable research skills in study design, execution and dissemination including publications and science communication.
3. create a collaborative culture of developing innovative medical solutions across Africa.
4. contribute to the open-science efforts of the global AI imaging community.
5. train and retain experts who can sustainably advance imaging research across Africa.
6. win medical imaging Grand Challenges!

To be eligible for the SPARK Academy, applicants must meet the following criteria:

1. Must be a resident in radiology, neurosurgery, or neurooncology, or a graduate student in medical physics, biomedical engineering, or computer science at an African institution.
2. For clinicians, it is recommended to have clinical or research experience with brain tumors.
3. For data science trainees, it is recommended to have intermediate to advanced programming experience.
4. Must be able to attend the on-site hackathon in person at one of the ten sites for all five days.
5. A self-motivated team player.

This project is one of the steps currently being taken by the Consortium for Advancement of MRI Education and Research in Africa (CAMERA) to create an unbroken chain of resources and expertise that will enable Africa to use MRI to realize their healthcare needs.

Prof. Udunna Anazodo at the Montreal Neurological Institute, McGill University, Montreal, Canada is the Director and Chair of the SPARK Academy and CAMERA, respectively. Dr. Abdul Nashirudeen Mumuni at the UDS-DIRC is the Site Director and Representative of the SPARK Academy and CAMERA in Ghana, respectively.