[Eeglablist] CUDAICA: GPU Optimization of Infomax-ICA EEG Analysis

[Juan Kamienkowski]

Dear eeglablist,

We recently developed a GPU-based solution for Infomax ICA
implementation (CUDAICA) that, comparing with the original BLAS and
CUBLAS versions, obtains a 25x increase of performance for the ICA
calculation. This is a couple of hours instead a couple of days of
processing. This is now published online in:

Federico Raimondo, Juan E. Kamienkowski, Mariano Sigman, and Diego
Fernandez Slezak "CUDAICA: GPU Optimization of Infomax-ICA EEG
Analysis" Computational Intelligence and Neuroscience. Volume 2012
(2012), Article ID 206972, 8 pages. doi:10.1155/2012/206972

http://www.hindawi.com/journals/cin/2012/206972/

We will be very glad if you want to try it. CUDAICA is freely
available from our wiki
(http://calamaro.exp.dc.uba.ar/cudaica/doku.php?id=start) with a
description of application features, FAQ and installation instructions
. CUDAICA works as a standalone application and integrates to the
EEGLAB Toolbox adding an option to process ICA using CUDAICA, just
like any other ICA implementation. It was designed for standard EEGLAB
users, with no extra effort needed to run this implementation. It
works under CUDA enabled hardware, that is, almost every modern
graphic card, making CUDAICA widely available and easy to use.

Best,

juan

-- 
Juan E Kamienkowski
Laboratorio de Neurociencia Integrativa
Departamento de Fisica, FCEN-UBA
Ciudad Universitaria, Pabellon I
(1428) Buenos Aires, Argentina
Phone: (54-11) 4576 3300 (282)
Fax: (54-11) 4576 3357
http://www.neurociencia.df.uba.ar/