Laboratory of Physiology of Cognitive Processes
2009
  • Title:SANDBOX, an interactive fMRI data visualization toolbox
  • Authors:S. Stoewer; J. Duncan; A. Bartels; G. A. Keliris; N. K. Logothetis; N. Sigala
  • Title of Journal:Cambridge Neuroscience Seminar for 2009
  • Year:2009
  • DOI:
Abstract
Conventional functional imaging software is often not optimized for immediate feedback on the results of analysis procedures during processing. Various preprocessing steps can be applied, but functional maps are only visible after all steps are completed. In cases where a rapid preview of activity maps is useful, e.g. in the development of new scanning sequences or in testing new experimental parameters, it is desirable to be able to interact with the data more directly and get results immediately. We have developed a toolbox that allows fast preview of activity and enables fast evaluation of scanning or experimental parameters between scans. This Matlab based toolbox allows the user to read, display and manipulate MRI/fMRI data into a graphical user interface. Resulting changes in functional maps can be observed while applying various preprocessing steps. Hemodynamic predictors can be generated from the onsets of known events, single and multiple trials can then be selected and analyzed during a scanning se ssion. Volumes and trials containing artifacts can be selected and removed from the analysis. Various processing steps (e.g. brain extraction, realignment, smoothing) and various filters can be applied. Three different models (Correlation, General Linear Model and Robust Regression) can be used to view resulting activation maps. Thresholds can be changed with a slider with instantaneous display of the changes. Additionally, diagnostic maps can be displayed. The software reads and writes standard ‘Analyze’ format files, as well as a proprietary format that additionally stores experimental design and possible eye movement and motion sensor data. This toolbox is not intended to replace tried and tested software, but to give the user the possibility to interactively visualize data processing steps and results before running established imaging software.