Multidimensional Scaling of Brain Imaging data key words: rsfMRI data, connectivity matrix, multidimensional scaling. This code covers multidimensional scaling approach for analysis of proximities (similiarities or dissimilarities) that reaveals structure and facilitates visualization of high dimensional data. MDS can be used as a preprocessing step for dimensionality reduction in classification and regression problems to find a lower dimensional representation for a set of objects (e.g., stimuli, brain regions, individuals, etc.) by representing the interobject proximities as distances in some lower dimensional space. This code includes:
- Extract time series
- Functional Brain Parcellation using Schaefer (2018) atlas (100 Rois, 7 Networks)
- Multiple subjects correlation matrices (4 datasets)
- Sklearn.manifold.MDS was applied
- Graph Representation
Other useful links to MDS:
https://github.com/stober/mds https://github.com/fredcallaway/brain_matrix https://github.com/drewwilimitis/hyperbolic-learning https://github.com/GeostatisticsLessons/GeostatisticsLessonsNotebooks/blob/master/notebooks/mds/mds.ipynb https://github.com/swethapola/Multidimensional-Scaling https://towardsdatascience.com/mds-multidimensional-scaling-smart-way-to-reduce-dimensionality-in-python-7c126984e60b https://stackabuse.com/guide-to-multidimensional-scaling-in-python-with-scikit-learn/ http://www.nervouscomputer.com/hfs/cmdscale-in-python/