Module 39-Inf-BDS Biomedical Data Science for Modern Healthcare Technology

Non-official translation of the module descriptions. Only the German version is legally binding.

The lecture puts the emphasis on data science techniques that are of actual relevance in biomedicine and health care. Important examples are the efficient administration of great amounts of patient data and their safe sharing, the involvement of of individual genetic patient records in clinical decision support processes, or the clinically effective evaluation of wearables. The seminar focuses on understanding, interpreting and evaluating scientific literature relating to topics treated in the lecture: the identification of issues that are critical in terms of data science, understanding how solutions can be developed, and understanding the (likely tricky) solutions themselves; last, gauging the range of open questions that remain. In the presentation, we will train presentation techniques (e.g. ‘why-what-how’), how to reach the audience, and how to deliver take-home messages that the audience can keep in mind. Writing a scientific report is meant to practice how to draft text that complies with scientific and ethical standards. A relevant aspect in that is reproducibility.

Contents of the lecture and the seminar touch upon topics that are rooted in data science (such as ‘deep learning’, ‘blockchain’ or graph databases) that play an ever more crucial, if not decisive role in modern biomedicine and healthcare.
Further examples are artificial intelligence, and here in particular (deep) machine learning for targeted clinical decision support, the aforementioned blockchain techniques (which, in other words, reflect bitcoin techniques applied for safe and efficient sharing of ethically critical patient data), genomics, internet of things, robotics, graph databases, ambient computing and visualization. For actual reasons, epidemiological methodology for analysis and control of epidemics can play an important role as well.

English version basic knowledge of algorithms, data structures and artificial intelligence are required. Moreover, depending on the particular topics covered, basic knowledge of machine learning, big data analysis, DNA/RNA sequencing and fundamental statistics are useful.

—