Thien Bao Nguen joins the SPL lab and Golby Lab in collaboration with Lauren O'Donnell to work on the supervised tract segmentation using groupwise registration rather than image based method.
You are here
The accepted paper with title "User-centric Affective Video Tagging from MEG and Peripheral Physiological Responses" (by M. Khomami Abadi, S. Mostafa Kia, R. Subramanian, P. Avesani, N. Sebe) is presented by Mojtaba Khomami Abadi and Ramanathan Subramanian in ACII 2013-Geneva.
NILab takes part as a working package to ATTEND project, a framework involving joint efforts from several other local research actors. Goal of the ATTEND project is to characterize and improve brain mechanisms of attention. NILab contribution will be mainly to extract shared patterns of attention between participants developing on - purpose original data analysis methods.
Diana Porro joins NILab as a Post-Doc to work on a research project on diffusion magnetic resonance imaging and pattern recognition. Diana received her diploma of Computer Science Engineer from the Technical University "José A. Echeverría" (CUJAE), Havana, Cuba in 2007. From September 2007 to November 2012, she worked as a researcher and software developer in the Advanced Technologies Application Center, Cuba. During this period she also enrolled in a PhD with the Pattern Recognition Laboratory of TUDelft, Netherlands.
Elena Kalinina joins NILab as a PhD student, XXIX cycle, ICT School UniTn
Mostafa Seyed Kia joins NILab as a PhD student, XXIX cycle, ICT School UniTn
The paper titled "Multiple-scale Visualization of Large Data based on Hierarchical Clustering" is accepted at the 2013 6th International Conference on Computer Science and Information Technology (ICCSIT 2013). The paper also will be published in IJCEE (International Journal of Computer and Electrical Engineering).
The School was held at the Department of Informatics and Mathematical Modelling, Technical University of Denmark, Section of Cognitive Systems (Copenhagen, Denmark) on August 12-16th, 2013. The topics covered included: introduction to Bayesian Inference, probabilistic regression, introduction to the analysis fo learning algorithms, Bayesian methods for sparse modelling of signals and images, non-parametric Bayesian modeling of complex networks, learning from small samples in high dimensions, fast, accurate and parameter-free Markov chain Monte Carlo for latent Gaussian models.
Alessandro Lopopolo joins NILab for his internship/thesis, Master in Cognitive Neuroscience Program, UniTn
In the paper, the discrete cosine transform (DCT) coefficients are proposed as a new and effective set of features for recognizing patterns of brain activity in MEG recording. Classification results on single-trial MEG decoding suggest that DCT is an efficient technique to reduce the dimensionality of feature space without losing discriminative power in brain decoding tasks.