You are here



The paper 'Brain decoding via graph kernels' was accepted for publication in the proceedings of the PRNI 2013 conference. The main contribution of this paper is the introduction of a graph kernel with a lower computational complexity that allows taking advantage from both the representative power of graphs and the discrimination power of kernel-based classifiers such as Support Vector Machines.


The paper "Fast Clustering for Interactive Tractography" (by Emanuele Olivetti, Thien Bao Nguyen, Eleftherios Garyfallidis, Nivedita Agarwal, Paolo Avesani) is accepted for presentation at the 3rd PRNI conference


The authors present a novel approach to the brain decoding problem: a high-dimensional two-sample test to determine whether the set of brain recordings related to one kind of stimulus, i.e. the first sample, and the ones related to the other kind of stimulus, i.e. the second sample, are drawn from the same probability distribution or not. The MEG data used was a Face, House and Body discrimination task.


Seyed Mostafa Kia is awarded master degree in cognitive neuroscience from University of Trento. His master thesis was on "Mass-Univariate Hypothesis Testing on MEG Data using Cross-Validation".


The mobility grant is aimed at providing financial support to do internship in a foreign country. Through this program, TrentoRise want to broaden the network with university/institute/organization around the world for sharing research's idea, transferring techniques. Thien Bao Nguyen got the grant for his internship at Harward Medical School, USA


The 3rd International Workshop on Pattern Recognition in Neuroimaging (PRNI 2013) was held in Philadelphia from June 22nd to 24th. The scientific program of the conference featured a wide range of presentations and posters focusing on functional and structural multivariate image analysis and machine learning methods. Sandro Vega-Pons & Paolo Avesani presented the poster "Brain decoding via graph kernels".


The poster presented was entitled "The Kernel Two-Sample Test vs. Brain Decoding", and the paper with the same title was accepted for publication in the conference proceedings. in the paper, the kernel two-sample test is proposed as an alternative to the use of classifiers for decoding stimuli presented to the subject. Its efficacy was tested on MEG data collected from a Body/Face/House discrimination task.


The accepted paper with title "Discrete Cosine Transform for MEG Signal Decoding" is presented in poster session of the 3rd  International Workshop on Pattern Recognition in Neuroimaging in Philadelphia.


The title of the presentation was "Fast Clustering for Interactive Tractography Segmentation". The paper with the same title was accepted for publication in the conference proceedings.


The paper "Clustering ensemble on reduced search spaces" was one of the 11 accepted papers on the ECML-PKDD workshop: 'Solving Complex Machine Learning Problems with Ensemble Methods' (COPEM). The paper presents some theoretical results that allows a dramatic prune of the search space for the median partition problem, therefore reducing the complexity some clustering ensemble algorithms.