Advancing data science research education in Africa through datathon-driven innovations

Area of Science:

• Biomedical Data Science
• Public Health Research
• Africa Health Informatics

Background:

• Growing volume of African biomedical data necessitates advanced data science skills for intervention development.
• International research initiatives generate substantial datasets requiring specialized analytical expertise.

Purpose of the Study:

• To present an innovative two-phase data science training model tailored for African researchers.
• To enhance the capacity for analyzing complex biomedical datasets and driving health discoveries in Africa.

Main Methods:

• A two-phase training model integrating foundational data science, machine learning, and traditional statistics.
• Utilizing multimodal data warehouses from completed research projects for hands-on analysis.
• Conducting an inaugural datathon with participants from 14 African countries focused on malaria research.

Main Results:

• Trainees engaged in group and competition-based learning, applying methods to real-world biomedical data.
• Case studies demonstrated the application of statistical and machine learning methods in regression and image analysis.
• Each datathon team initiated a research project, aiming for scientific manuscript publication.

Conclusions:

• The developed datathon learning model effectively advances data science skills among African researchers.
• This framework fosters multidisciplinary research and maximizes the impact of biomedical data resources for health interventions.
• The model serves as a comprehensive resource for implementing similar data science training programs globally.