Learn More
These notes give a short introduction to Gaussian mixture models (GMMs) and the Expectation-Maximization (EM) algorithm, first for the specific case of GMMs, and then more generally. These notes assume you're familiar with basic probability and basic calculus. If you're interested in the full derivation (Section 3), some familiarity with entropy and KL(More)
These notes give a short review of Hidden Markov Models (HMMs) and the forward-backward algorithm. They're written assuming familiarity with the sum-product belief propagation algorithm, but should be accessible to anyone who's seen the fundamentals of HMMs before. The notation here is borrowed from Introduction to Probability by Bertsekas & Tsitsiklis:(More)
We present an analysis framework for large studies of multimodal clinical quality brain image collections. Processing and analysis of such datasets is challenging due to low resolution, poor contrast, mis-aligned images, and restricted field of view. We adapt existing registration and segmentation methods and build a computational pipeline for spatial(More)
We develop a method for unsupervised analysis of functional brain images that learns group-level patterns of functional response. Our algorithm is based on a generative model that comprises two main layers. At the lower level, we express the functional brain response to each stimulus as a binary activation variable. At the next level, we define a prior over(More)
Functional MRI studies have uncovered a number of brain areas that demonstrate highly specific functional patterns. In the case of visual object recognition, small, focal regions have been characterized with selectivity for visual categories such as human faces. In this paper, we develop an algorithm that automatically learns patterns of functional(More)
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to(More)
We present a model that describes the structure in the responses of different brain areas to a set of stimuli in terms of stimulus categories (clusters of stimuli) and functional units (clusters of voxels). We assume that voxels within a unit respond similarly to all stimuli from the same category, and design a nonparametric hierarchical model to capture(More)
We propose and demonstrate an inference algorithm for the automatic segmentation of cerebrovascular pathologies in clinical MR images of the brain. Identifying and differentiating pathologies is important for understanding the underlying mechanisms and clinical outcomes of cerebral ischemia. Manual delineation of separate pathologies is infeasible in large(More)
We present an interactive tool for visualization of medical imaging pipelines that are built with Nipype, a freely available tool for building pipelines programatically. Our tool enables researchers to interact with their pipelines, visualize the pipeline structure, and view their intermediate and final results. We also provide a video and live(More)
The effective translation of data into novel insights, discoveries , and solutions, also known as data science, has enormous potential to bring about positive social change. In this paper, we propose ways to " democratize data science " : that is, to allocate the power of data science to society's greatest needs. Two underlying challenges are 1) the(More)