Mathematics for biomedical data analysis

Selected mathematical perspectives to introduce topical conversations and developments in the field. Algorithms: A set of simple mathematical concepts is used to classify a wide variety of computational techniques (logistic regression, principal component analysis, kernel methods, random forests, graphical models, deep learning) depending on their learning, versatility and application type, for a non-technical overview of the field. Inference: Covers the major mathematical principles underpinning data-dredging and related pitfalls, such as misinterpretation of P-values and multiple testing problems. Causality: Considers the mathematical basis for the paradigmatic shifts from traditional data analysis to causal analysis. Provoking examples are provided to 1) distinguish causal and non-causal relationships and 2) explain why causal inference is beyond the scope of the mathematical language of classical statistics.

Contents can vary.