PRoNTo Pattern Recognition for Neuroimaging Toolbox该模式识别的神经影像学工具箱.ppt

* * * * * * * * * * * * * * * * * * * * * * UFABC, Sao Paulo, Brazil Pattern Recognition MethodsPart 1: Kernel Methods SVM London, UK 21 May 2012 Janaina Mourao-Miranda, Machine Learning and Neuroimaging Lab, University College London, UK Outline Pattern Recognition: Concepts Framework Kernel Methods for Pattern Analysis Support Vector Machine (SVM) Pattern recognition aims to assign a label to a given pattern (test example) based either on a priori knowledge or on statistical information extracted from the previous seen patterns (training examples). The patterns to be classified are usually groups of measurements or observations (e.g. brain image), defining points in an appropriate multidimensional vector space. Types of learning procedure: Supervised learning: The training data consist of pairs of input (typically vectors, X), and desired outputs or labels (y). The output of the function can be a continuous (regression) or a discrete (classification) value. Unsupervised learning: The training data is not labeled. The aims is to to find inherent patterns in the data that can then be used to determine the correct output value for new data examples (e.g. clustering). Semi-supervised learning, reinforcement learning. Pattern Recognition Concepts Currently implemented in PRoNTo Machine Learning Methodology Testing Phase Prediction Learning/Training Phase Generate a function or classifier f such that Training Examples: (X1, y1), . . .,(Xs, ys) Test Example Xi f(xi) - yi f(Xi) = yi f Computer-based procedures that learn a function from a series of examples Pattern Recognition Framework Input (brain scans) X1 X2 X3 Output (control/patient) y1 y2 y3 No mathematical model available Standard Statistical Analysis (mass-univatiate) ... Voxel-wise GLM model estimation Independent statistical test at each voxel Correction for multiple comparisons Univariate Statistical Parametric Map Input Output Time BOLD signal Volumes from task 1 Volu