A General and Unifying Framework for Feature Construction, in Image-Based Pattern Classification

Section of Biomedical Image Analysis, Radiology Department, University of Pennsylvania, Philadelphia, PA 19014, USA.
Information processing in medical imaging: proceedings of the ... conference 02/2009; 21:423-34. DOI: 10.1007/978-3-642-02498-6_35
Source: PubMed


This paper presents a general and unifying optimization framework for the problem of feature extraction and reduction for high-dimensional pattern classification of medical images. Feature extraction is often an ad hoc and case-specific task. Herein, we formulate it as a problem of sparse decomposition of images into a basis that is desired to possess several properties: 1) Sparsity and local spatial support, which usually provides good generalization ability on new samples, and lends itself to anatomically intuitive interpretations; 2) good discrimination ability, so that projection of images onto the optimal basis yields discriminant features to be used in a machine learning paradigm; 3) spatial smoothness and contiguity of the estimated basis functions. Our method yields a parts-based representation, which warranties that the image is decomposed into a number of positive regional projections. A non-negative matrix factorization scheme is used, and a numerical solution with proven convergence is used for solution. Results in classification of Alzheimers patients from the ADNI study are presented.

Full-text preview

Available from:
  • Source
    • "Baseline MRI: AD vs. HC 87.9% pMCI vs. HC 83.2% pMCI vs. sMCI 70.4% Longit MRI: AD vs. HC 90.3% pMCI vs. HC 86.9% pMCI vs. sMCI 82.1% [13] 202 AD, 410 MCI, 236 HC 75% of data in training set: AD vs. HC 78.4% MCI vs. HC 71.2% 90% of data in training set: AD vs. HC 85.7% MCI vs. HC 79.2% [2] 56 AD, 60 MCI, 60 HC AD vs. HC 89% MCI vs. HC 72% [12] 198 AD, 238 sMCI, 167 pMCI, 234 HC AD vs. HC 88.8% sMCI vs. pMCI 69.6% Figure 3: Examples of convolutions with the fourth basis of the 3D sparse autoencoder (32nd slice). An example from each class is randomly chosen. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Pattern recognition methods using neuroimaging data for the diagnosis of Alzheimer's disease have been the subject of extensive research in recent years. In this paper, we use deep learning methods, and in particular sparse autoencoders and 3D convolutional neural networks, to build an algorithm that can predict the disease status of a patient, based on an MRI scan of the brain. We report on experiments using the ADNI data set involving 2,265 historical scans. We demonstrate that 3D convolutional neural networks outperform several other classifiers reported in the literature and produce state-of-art results.
    Full-text · Article · Feb 2015
  • Source
    • "Through employing pattern classification methods, neuroimaging has demonstrated its effectiveness in predicting Alzheimer's disease (AD) status based on individual magnetic resonance imaging (MRI) and/or positron emission tomography (PET) scans [5] [11] [18]. Because AD is a neurodegenerative disorder characterized by progressive impairment of memory and other cognitive functions, it is important to understand how structural and functional changes * Data collection and sharing for this project was funded by the "
    [Show abstract] [Hide abstract]
    ABSTRACT: Alzheimer's disease (AD) is a neurodegenerative disorder characterized by progressive impairment of memory and other cognitive functions, which makes regression analysis a suitable model to study whether neuroimaging measures can help predict memory performance and track the progression of AD. Existing memory performance prediction methods via regression, however, do not take into account either the interconnected structures within imaging data or those among memory scores, which inevitably restricts their predictive capabilities. To bridge this gap, we propose a novel Sparse Multi-tAsk Regression and feaTure selection (SMART) method to jointly analyze all the imaging and clinical data under a single regression framework and with shared underlying sparse representations. Two convex regularizations are combined and used in the model to enable sparsity as well as facilitate multi-task learning. The effectiveness of the proposed method is demonstrated by both clearly improved prediction performances in all empirical test cases and a compact set of selected RAVLT-relevant MRI predictors that accord with prior studies.
    Full-text · Conference Paper · Nov 2011
  • Source
    • "Pair-wise similarity between subjects encodes relationship between labeled and unlabeled data and it is shown to improve classification accuracy in presence of unlabeled data [7]. Finally, the method is cast as a constrained optimization problem similar to [1] but the optimization cost function and its constraints as well as our optimizer are significantly different. "
    [Show abstract] [Hide abstract]
    ABSTRACT: We present a new semi-supervised algorithm for dimensionality reduction which exploits information of unlabeled data in order to improve the accuracy of image-based disease classification based on medical images. We perform dimensionality reduction by adopting the formalism of constrained matrix decomposition of to semi-supervised learning. In addition, we add a new regularization term to the objective function to better captur the affinity between labeled and unlabeled data. We apply our method to a data set consisting of medical scans of subjects classified as Normal Control (CN) and Alzheimer (AD). The unlabeled data are scans of subjects diagnosed with Mild Cognitive Impairment (MCI), which are at high risk to develop AD in the future. We measure the accuracy of our algorithm in classifying scans as AD and NC. In addition, we use the classifier to predict which subjects with MCI will converge to AD and compare those results to the diagnosis given at later follow ups. The experiments highlight that unlabeled data greatly improves the accuracy of our classifier.
    Full-text · Conference Paper · May 2011
Show more