Message from Program Chairs: PRNI 2015

Pattern recognition techniques have become a staple of neuroimaging data analysis, with a growing number of dedicated open source toolboxes readily available to the researchers in neuroscience. In parallel, there has been a large increase in the availability of public datasets. Not only has the amount of data increased, but the variety of modalities has also kept up, with behavioral test results, biomarker data, genotype and gene expression data now becoming abundant. This has created new challenges for data storage and distribution, and emphasized the critical role of state-of-the-art multivariate and multimodal techniques. The dialogue between producers and consumers of multivariate, predictive modeling methods enabled by PRNI is ongoing, with great benefits for all involved. We believe that the high quality and breadth of the program is evidence of the groundbreaking research at the intersection of pattern recognition and neuroimaging. Submitted manuscripts spanned multiple areas including sparse techniques, graph embedding, graphical models, and multimodal integration. Further, we observed a growing interest in statistical techniques for scientific interpretation purposes beyond basic decoding. In particular, we are delighted to see a number of submissions tackling the open problem of identifying statistically significant features from classifier weights. Twenty four manuscripts were selected for publication after an extensive review by at least two members of the program committee and thorough discussions among the program chairs. This year also marked the first call for late breaking abstracts designed to highlight promising work in progress, and presented as posters.