A teaching proposal for a short course on biomedical data science

Area of Science:

• Biomedical Data Science
• Computational Statistics
• Machine Learning

Background:

• Increasing volume of big biomedical data necessitates specialized training for university students.
• Existing curricula may not adequately cover the practical aspects of analyzing and interpreting health data.

Purpose of the Study:

• To propose and describe a master's degree course plan for biomedical data science.
• To share practical experiences from implementing the course in the last academic year.

Main Methods:

• Curriculum development focusing on data acquisition, cleaning, and preparation.
• Instruction on exploratory data analysis (EDA), machine learning (supervised/unsupervised), and result validation.
• Integration of open science principles, utilizing open-source tools and data.

Main Results:

• Students were trained to identify, clean, and prepare open biomedical datasets.
• Practical application of statistical and machine learning techniques with interpretation of outcomes.
• Emphasis on using open-source programming languages (R, Python) and open-access resources.

Conclusions:

• The proposed curriculum effectively equips students with essential biomedical data science skills.
• The course promotes best practices in data analysis and interpretation within an open science framework.
• This teaching proposal serves as a valuable resource for developing new biomedical data science programs.