Multivariate Effect Ranking via Adaptive Sparse PLS

Conference Paper · June 2015with12 Reads
DOI: 10.1109/PRNI.2015.27
Conference: Conference: 2015 International Workshop on Pattern Recognition in NeuroImaging (PRNI)
Unsupervised learning approaches, such as Sparse Partial Least Squares (SPLS), may provide useful insights into the brain mechanisms by finding relationships between two sets of variables (i.e. Views) from the same subjects. The algorithm outputs two sets of paired weight vectors, where each pair expresses an "effect" between both views. However, each effect can be described by a different number of variables. In this paper, we propose a novel approach to find multivariate associations between combinations of clinical/behavioural variables and brain voxels/regions which provides an unique solution with different levels of sparsity per weight vector pair. The effects described by the weight vector pairs are ranked by how much data covariance they explain. The proposed method was able to find statistically significant effects or relationships in a dementia dataset between clinical/demographic information and brain scans. Its adaptive nature allowed not only to determine an optimal sparse solution, but also provided the flexibility to select the adequate number of clinical/demographic variables and voxels to describe each effect, which enabled it to distinguish the effects associated with age from the ones associated with dementia.
The aims of the project are to ) develop an ontology of behaviour change interventions b) use artificial intelligence to annotate and synthesise the published literature using the ontology and c) d…" [more]
We have developed structured prediction methods for predicting output objects with graphical structure, which may be given as prior information, drawn randomly, or inferred during learning the mode…" [more]
Conference Paper
September 2016
    Sparse Canonical Correlation Analysis (SCCA) has been proposed to find pairs of sparse weight vectors that maximize correlations between sets of paired variables. This is done by computing one weight vector pair, deflating the correlation matrix between the views, and then repeating the process to compute the next pair. However, the deflation step used does not guarantee the orthogonality of... [Show full abstract]
    June 2016 · Journal of Neuroscience Methods · Impact Factor: 2.05
      Background: Supervised classification machine learning algorithms may have limitations when studying brain diseases with heterogeneous populations, as the labels might be unreliable. More exploratory approaches, such as Sparse Partial Least Squares (SPLS), may provide insights into the brain's mechanisms by finding relationships between neuroimaging and clinical/demographic data. The... [Show full abstract]
        Recently machine learning methodology has been used increasing to analyze the relation-ship between stimulus categories and fMRI responses [2, 14, 15, 11, 13, 8, 9, 1, 12, 7]. Here, we introduce a new unsupervised machine learning approach to fMRI analysis ap-proach, in which the simple categorical description of stimulus type (e.g. type of task) is replaced by a more informative vector of... [Show full abstract]
        Discover more