Datasets
Standard Dataset
Locally Linear Embedding and fMRI feature selection in psychiatric classification
- Citation Author(s):
-
SidhuGagan
University of Alberta - Submitted by:
- Gagan Sidhu
- Last updated:
- Tue, 05/17/2022 - 22:17
- DOI:
- 10.21227/zkkm-es92
- Data Format:
- Research Article Link:
- Links:
- License:
851 Views- Categories:
- Keywords:
Abstract
This dataset collection contains eleven datasets used in Locally Linear Embedding and fMRI feature selection in psychiatric classification.
The datasets given in the Links section are reduced subsets of those contained in their respective tar files (a consequence of Mendeley Data's 10GB limitation).
The Linked datasets (not the tar files) contain just the MATLAB file and the resting state image (or block-design fMRI for the MRN dataset), where appropriate.
The datasets in the tar file contain all of the preprocessing intermediates for the fMRI input to the LLE algorithm.
Information regarding data acquisition for the datasets used in this article can be found here, and here.
Please consult 'instructions' for the contents of each tar file.
Please consult 'Cracking the code to unlock the human brain'(published originally as an Opinion piece in the Edmonton Journal on 7 November 2007) to learn more about Dr. Sam Roweis' (1972-2010) belief on the benefits of correctly applying the LLE algorithm.
Dataset Files
- COBRE.tar (20.52 GB)
- MRN.tar (67.82 GB)
- Peking_1.tar (45.74 GB)
- Peking_2.tar (40.16 GB)
- Peking_3.tar (23.37 GB)
- UM_1.tar (66.55 GB)
- UM_2.tar (25.18 GB)
- UCLA_1.tar (23.47 GB)
- Yale.tar (22.88 GB)
- Stanford.tar (18.41 GB)
- USM.tar (66.13 GB)
Documentation
| Attachment | Size |
|---|---|
 Dataset nomenclature. | 63.22 KB |
| Cracking the code to unlock the human brain | 50.11 KB |
