An introduction to Dr Mark Griffin – a data scientist working with the Auckland Bioengineering Institute and the Mātai Medical Research Institute

Mark is a data scientist based in the Gold Coast Health and Knowledge Precinct Australia. He works with the Mātai Medical Research Institute in Gisborne (as an Industry Senior Research Fellow), and as a result with people from the ABI (particularly with Dr Vickie Shim). He expects to be visiting the ABI and Mātai two or three times a year in future. On this first visit Mark will give an overview of the broad range of methodologies that he is working with covering both statistics and mathematical modelling for magnetic resonance imaging (MRI). Mark is probably the most active presenter of short courses in statistics in Australia having now presented over 100 two-day and 50 five-day workshops. This one-hour seminar is a heavily condensed version of approximately 40 days of workshop material that he regularly presents. On future visits Mark will talk about more focussed areas of his work. In this whirlwind talk Mark will cover: • Regression – the difference between linear, logistic and Poisson regression • Regression assumptions – why can’t I just enter the names of my variables into the software and click “run”? • Factor analysis – what is factor analysis, what results are produced, what is rotation, and what is the difference between factor analysis and principal component analysis? • Cluster analysis – what is cluster analysis? What is the difference between k means and hierarchical clustering? • Longitudinal data analysis – how mixed effects models account for the temporal correlation in the data • Survival or time-to-event analysis – what is survival analysis? • Missing data – why can’t I just ignore the observations with missing data? How does multiple imputation help? • Structural Equation Modelling – what is SEM? • Data visualization – why do I need to invest time customizing my data visualization? • Geographical Information Systems (GIS) – what is GIS? • Survey design – why do survey questions need to be worded carefully? • Applications in imaging processing for magnetic resonance imaging – Finite Element Models, filtering • Data management for complex clinical trials