Big Data Analytics and Machine Learning – Biostatistics and Epidemiology in Translational Medicine

Upon completing the course, participants should be able to:

Knowledge

• Describe the benefits and limitations of statistical analyses based on experimental and non-experimental data sources; particularly, list and distinguish common biases and pitfalls in the analysis of observational studies.
• Explain the concept of a "question-first" approach to statistical analysis, and particularly the use of a causal language to frame and communicate scientific questions and identify examples of research questions that could be addressed.
• Summarise the advantages and limitations of using machine learning tools in medical research and explain the overall differences between prediction and inference for interpretable parameters.

Skills

• Employ causal inference tools and target experiment conceptualisation to translate relevant scientific questions into well-defined statistical parameters in statistical collaborations.
• Utilise causal diagrams to discuss current scientific knowledge, and to identify and evaluate potential sources of bias.
• Apply critical thinking in evaluation of scientific literature and when engaging in scientific collaborations.
   

Competences

• Communicate and collaborate more effectively with subject matter experts and professional statisticians to answer pressing research questions in medicine based on big data and machine learning.
• Evaluate the benefits and limitations of (causal) statistical analyses incorporating machine learning tools.
• Recognise the ethical implications of big data and machine learning in translational medicine and be able to engage in discussions on responsible use and interpretation of these tools.