This is a experimental version of python code that can be used instead of or in addition to (Matlab-based) BodySPM by Enrico Glerean (a version of matlab-based BodySPM is available at https://version.aalto.fi/gitlab/eglerean/sensations). It is designed to work with data coming from a version of https://version.aalto.fi/gitlab/eglerean/embody .
- read in the data from theonline data collection system, and converts the topographies to .csv (with each cell representing one pixel).
- additional subject info and info about the stimuli is saved as JSONs. (Csv and JSON should be suitable for long term data storage and cross-platform compatibility)
- combine data from multiple subjects into one hdf5 file for analysis
- pixel-wise one sample t-test
- pixel-wise comparison of two groups, either using two sample t-test or z test of proportions
- count pixels within a given mask (e.g. within body outline)
- correlate maps with background variables (e.g. age)
- draw single subject colouring within mask
- draw single subject colouring as-collected (useful for visual quality control)
- easy multiple comparison correction to use with pixel-wise analyses
- binarise maps to coloured/not coloured
- Region of Interest analyses (currently exists as a script, to be converted to a function)
- correlate two maps with one another / other similarity measure
- potentially functions for visualising analysis outputs? (currently all except single subject visualisations are done as scripts)