Skip to content

Latest commit

 

History

History
27 lines (18 loc) · 1.39 KB

README.md

File metadata and controls

27 lines (18 loc) · 1.39 KB
Raw

SparsEDA Python

SparsEDA is an electrodermal activity (EDA, i.e., galvanic skin response, GSR) decomposition method. This code is translated from the folliwng github page: https://github.com/fhernandogallego/sparsEDA

Following figure shows and performance compared to the original code written for Matlab (five females and five males).

Ten subjects (Five females and five males) with pain and stress data have been tested to compare with the original code written in Matlab).

Following figure shows the length (sec) distribution of 10 subjects:

Following figure shows the performance distribution between this and Matlab codes:

  • Please resample to ~8 Hz if sampling rate is greater than 8 Hz.
  • Tested with Python 3.6.9, SciPy 1.3.1, and Numpy 1.19.5.

Author information: Biomedical Engineering, University of Connecticut

If you use this code, please cite these for your literature works:

  1. Hernando-Gallego, Francisco, David Luengo, and Antonio Artés-Rodríguez. "Feature extraction of galvanic skin responses by nonnegative sparse deconvolution." IEEE journal of biomedical and health informatics 22.5 (2017): 1385-1394.
  2. Hernando-Gallego, Francisco, David Luengo, and Antonio Artés-Rodríguez. https://github.com/fhernandogallego/sparsEDA
  3. Kong, Youngsun, https://github.com/yskong224/SparsEDA-python