Statistical methods and analysis for large datasets
SP-0985
Abstract
Statistical methods and analysis for large datasets
1OncoRay – National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden - Rossendorf, Dresden, Germany
Show Affiliations
Hide Affiliations
Abstract Text
In this talk, methods will be presented that are able to deal with big datasets. For that purpose, we will discuss a data analysis workflow from data pre-processing to final model interpretation, including relevant aspects such as unsupervised data aggregation, analysis of batch effects, normalisation, training-validation-test approaches, feature selection, hyperparameter tuning, and model interpretation. The methodological steps will be illustrated with examples from radiomics or transcriptomics.