Statistical Parametric Mapping

Project: Other

Project Details


IMI Academic Visitor and Department for Health Improving Human Function funds

This application aims to support a one week visit from Dr Todd Pataky, who is an outstanding scholar in the field of analysis and simulation of random processes in biomechanical systems. Over the last years Dr Pataky has developed Statistical Parametric Mapping (SPM) methods for human movement sciences and has actively promoted their free dissemination and application in multiple areas of biomechanics (see, for example, SPM overcomes some limitations of conventional analysis approaches such as continuum discretization and ad dimensionality reduction. SPM has been the gold standard in Neuroimaging for two decades, and will likely also become standard in human movement sciences and possibly other fields which deal with smooth continuous variables.
During Dr Pataky’s visit, we plan to carry out the following activities:
- A 3-day workshop open to staff and students, as well as University collaborators and partners. The workshop will be planned with an increasing level of complexity to provide different levels of learning outcomes and therefore target a broader range of interests (application and/or method oriented). Dr Pataky has an established track record of dissemination of SPM through workshops oriented to different cohorts (scientists, practitioners, etc.).
Day 1 - basic introduction to SPM, hands-on examples of how freely available tools could be used to apply hypothesis testing to continuous data, and individual consultations
Day 2 –more complex Mixed Models involving smooth continuous data, and an interactive session where participants can apply SPM to their own datasets
Day 3 - an in-depth explanation of the theory behind SPM and of how it has been implemented through Matlab coding
The discussion and development of possible collaborative initiatives, which will initially focus on the integration of SPM with a magnitude-based inferences approach, and the extension of the use of SPM to circular variables.

Statistical Parametric Mapping is a statistical technique that could be used to extend the use of Linear Mixed Modelling to sets of smooth continuous data. Although most of the applied research Dr Pataky has carried out is in the area of human movement, the potential of the method extends to cover many of the research areas present at the University of Bath (e.g. biomedical engineering, neuroscience, computer science, mathematics, biology). The workshop appears as the natural continuation of the recent workshop on Mixed Modelling and Magnitude Based Inferences delivered by Prof Will Hopkins and supported by IMI. The University will benefit by offering scientists and practitioners from different disciplines the opportunity to come together, discuss about common issues in the analysis of continuous timeseries and understand how Mixed Modelling approaches could be extended from discrete to continuous data. They will also have the opportunity to apply SPM to case-study scenarios and their own data.
The visit will also initiate a collaborative venture between our Universities and aim at improving current applied techniques in the area of human movement science. The outcomes of such collaboration will improve data processing and analysis techniques in the many experimental researches we carry out and will potentially impact a multitude of areas related to health and wellbeing, such as injury prevention, clinical biomechanics and sports performance.
Short title£3,000
Effective start/end date22/01/1727/01/17