A connection between pattern classification by machine learning and statistical inference with the General Linear Model

A connection between the general linear model (GLM) with frequentist statistical testing and machine learning (MLE) inference is derived and illustrated. Initially, the estimation of GLM parameters is expressed as a Linear Regression Model (LRM) of an indicator matrix; that is, in terms of the inverse problem of regressing the observations. Both approaches, i.e. GLM and LRM, apply to different domains, the observation and the label domains, and are linked by a normalization value in the least-squares solution. Subsequently, we derive a more refined predictive statistical test: the linear Support Vector Machine (SVM), that maximizes the class margin of separation within a permutation analysis. This MLE-based inference employs a residual score and associated upper bound to compute a better estimation of the actual (real) error. Experimental results demonstrate how parameter estimations derived from each model result in different classification performance in the equivalent inverse problem. Moreover, using real data, the MLE-based inference including model-free estimators demonstrates an efficient trade-off between type I errors and statistical power.

A Machine Learning Approach to Reveal the NeuroPhenotypes of Autisms

Although much research has been undertaken, the spatial patterns, developmental course, and sexual dimorphism of brain structure associated with autism remains enigmatic. One of the difficulties in investigating differences between the sexes in autism is the small sample sizes of available imaging datasets with mixed sex. Thus, the majority of the investigations have involved male samples, with females somewhat overlooked. This paper deploys machine learning on partial least squares feature extraction to reveal differences in regional brain structure between individuals with autism and typically developing participants. A four-class classification problem (sex and condition) is specified, with theoretical restrictions based on the evaluation of a novel upper bound in the resubstitution estimate. These conditions were imposed on the classifier complexity and feature space dimension to assure generalizable results from the training set to test samples. Accuracies above [Formula: see text] on gray and white matter tissues estimated from voxel-based morphometry (VBM) features are obtained in a sample of equal-sized high-functioning male and female adults with and without autism ([Formula: see text], [Formula: see text]/group). The proposed learning machine revealed how autism is modulated by biological sex using a low-dimensional feature space extracted from VBM. In addition, a spatial overlap analysis on reference maps partially corroborated predictions of the “extreme male brain” theory of autism, in sexual dimorphic areas.

Using deep neural networks along with dimensionality reduction techniques to assist the diagnosis of neurodegenerative disorders

The analysis of neuroimaging data is frequently used to assist the diagnosis of neurodegenerative disorders such as Alzheimer's disease (AD) or Parkinson's disease (PD) and has become a routine procedure in the clinical practice. During the past decade, the pattern recognition community has proposed a number of machine learning-based systems that automatically analyse neuroimaging data in order to improve the diagnosis. However, the high dimensionality of the data is still a challenge and there is room for improvement. The development of novel classification frameworks as TensorFlow, recently released as open source by Google Inc., represents an opportunity to continue evolving these systems. In this work, we demonstrate several computer-aided diagnosis (CAD) systems based on Deep Neural Networks that improve the diagnosis for AD and PD and outperform those based on classical classifiers. In order to address the small sample size problem we evaluate two dimensionality reduction algorithms based on Principal Component Analysis and Non-Negative Matrix Factorization (NNMF), respectively. The performance of developed CAD systems is assessed using 4 datasets with neuroimaging data of different modalities.

Ensemble of random forests One vs. Rest classifiers for MCI and AD prediction using ANOVA cortical and subcortical feature selection and partial least squares

BACKGROUND

Alzheimer's disease (AD) is the most common cause of dementia in the elderly and affects approximately 30 million individuals worldwide. Mild cognitive impairment (MCI) is very frequently a prodromal phase of AD, and existing studies have suggested that people with MCI tend to progress to AD at a rate of about 10-15% per year. However, the ability of clinicians and machine learning systems to predict AD based on MRI biomarkers at an early stage is still a challenging problem that can have a great impact in improving treatments.

METHOD

The proposed system, developed by the SiPBA-UGR team for this challenge, is based on feature standardization, ANOVA feature selection, partial least squares feature dimension reduction and an ensemble of One vs. Rest random forest classifiers. With the aim of improving its performance when discriminating healthy controls (HC) from MCI, a second binary classification level was introduced that reconsiders the HC and MCI predictions of the first level.

RESULTS

The system was trained and evaluated on an ADNI datasets that consist of T1-weighted MRI morphological measurements from HC, stable MCI, converter MCI and AD subjects. The proposed system yields a 56.25% classification score on the test subset which consists of 160 real subjects.

COMPARISON WITH EXISTING METHOD(S)

The classifier yielded the best performance when compared to: (i) One vs. One (OvO), One vs. Rest (OvR) and error correcting output codes (ECOC) as strategies for reducing the multiclass classification task to multiple binary classification problems, (ii) support vector machines, gradient boosting classifier and random forest as base binary classifiers, and (iii) bagging ensemble learning.

CONCLUSIONS

A robust method has been proposed for the international challenge on MCI prediction based on MRI data. The system yielded the second best performance during the competition with an accuracy rate of 56.25% when evaluated on the real subjects of the test set.

An Optimal Approach for Selecting Discriminant Regions for the Diagnosis of Alzheimer's Disease

In this work, we present a fully automatic computer-aided diagnosis method for the early diagnosis of the Alzheimer's disease. We study the distance between classes (labelled as normal controls and possible Alzheimer's disease) calculated in 116 regions of the brain using the Welchs's t-test. We select the regions with highest Welchs's t-test value as features to perform classification. Furthermore, we also study the less discriminative region according to the t-test (regions with lowest t-test absolute values) in order to use them as reference. We show that the mean and standard deviation of the intensity values in these two regions, the less and most discriminative according to the Welch's ttest, can be combined as a vector. The modulus and phase of this vector reveal statistical differences between groups which can be used to improve the classification task. We show how they can be used as input for a support vector machine classifier. The proposed methodology is tested in a SPECT brain database of 70 SPECT brain images yielding an accuracy up to 91.5% for a wide range of selected voxels.

Assessing polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans in air across Latin American countries using polyurethane foam disk passive air samplers

A passive air sampling network has been established to investigate polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) at Global Atmospheric Passive Sampling (GAPS) sites and six additional sites in the Group of Latin American and Caribbean Countries (GRULAC) region. The air sampling network covers background, agricultural, rural, and urban sites. Samples have been collected over four consecutive periods of 6 months, which started in January 2011 [period 1 (January to June 2011), period 2 (July to December 2011), period 3 (January to June 2012), and period 4 (July 2012 to January 2013)]. Results show that (i) the GAPS passive samplers (PUF disk type) and analytical methodology are adequate for measuring PCDD/F burdens in air and (ii) PCDD/F concentrations in air across the GRULAC region are widely variable by almost 2 orders of magnitude. The highest concentrations in air of Σ4-8PCDD/Fs were found at the urban site São Luis (Brazil, UR) (i.e., 2560 fg/m3) followed by the sites in São Paulo (Brazil, UR), Mendoza (Argentina, RU), and Sonora (Mexico, AG) with values of 1690, 1660, and 1610 fg/m3, respectively. Very low concentrations of PCDD/Fs in air were observed at the background site Tapanti (Costa Rica, BA), 10.8 fg/m3. This variability is attributed to differences in site characteristics and potential local/regional sources as well as meteorological influences. The measurements of PCDD/Fs in air agree well with model-predicted concentrations performed using the Global EMEP Multimedia Modeling System (GLEMOS) and emission scenario constructed on the basis of the UNEP Stockholm Convention inventory of dioxin and furan emissions.

Automatic assistance to Parkinson's disease diagnosis in DaTSCAN SPECT imaging

PURPOSE

In this work, an approach to computer aided diagnosis (CAD) system is proposed as a decision-making aid in Parkinsonian syndrome (PS) detection. This tool, intended for physicians, entails fully automatic preprocessing, normalization, and classification procedures for brain single-photon emission computed tomography images.

METHODS

Ioflupane[(123)I]FP-CIT images are used to provide in vivo information of the dopamine transporter density. These images are preprocessed using an automated template-based registration followed by two proposed approaches for intensity normalization. A support vector machine (SVM) is used and compared to other statistical classifiers in order to achieve an effective diagnosis using whole brain images in combination with voxel selection masks.

RESULTS

The CAD system is evaluated using a database consisting of 208 DaTSCAN images (100 controls, 108 PS). SVM-based classification is the most efficient choice when masked brain images are used. The generalization performance is estimated to be 89.02 (90.41-87.62)% sensitivity and 93.21 (92.24-94.18)% specificity. The area under the curve can take values of 0.9681 (0.9641-0.9722) when the image intensity is normalized to a maximum value, as derived from the receiver operating characteristics curves.

CONCLUSIONS

The present analysis allows to evaluate the impact of the design elements for the development of a CAD-system when all the information encoded in the scans is considered. In this way, the proposed CAD-system shows interesting properties for clinical use, such as being fast, automatic, and robust.

Improved parkinsonism diagnosis using a partial least squares based approach

PURPOSE

An accurate and early diagnosis of Parkinsonian syndrome (PS) is nowadays a challenge. This syndrome includes several pathologies with similar symptoms (Parkinson's disease, multisystem atrophy, progressive supranuclear palsy, corticobasal degeneration and others) which make the diagnosis more difficult. (123)I-ioflupane allows to obtain in vivo images of the brain that can be used to assist the PS diagnosis and provides a way to improve its accuracy.

METHODS

In this paper, we show a novel method to automatically classify (123)I-ioflupane images into two groups: controls or PS. The proposed methodology analyzes separately each hemisphere of the brain by means of a novel approach based on partial least squares (PLS) and support vector machine.

RESULTS

A database with 189 (123)I-ioflupane images (94 controls and 95 pathological images) was used for evaluation purposes. The application of the proposed method based on PLS yields high accuracy rates up to 94.7% with sensitivity = 93.7% and specificity = 95.7%, outperforming previous approaches based on singular value decomposition, which are used as a reference.

CONCLUSIONS

The use of advanced techniques based on classical signal analysis and their application to each hemisphere of the brain separately improves the (assisted) diagnosis of PS.

Feature selection using factor analysis for Alzheimer's diagnosis using 18F-FDG PET images

PURPOSE

This article presents a computer-aided diagnosis technique for improving the accuracy of the early diagnosis of Alzheimer's disease (AD). Two hundred and ten 18F-FDG PET images from the ADNI initiative [52 normal controls (NC), 114 mild cognitive impairment (MCI), and 53 AD subjects] are studied.

METHODS

The proposed methodology is based on the selection of voxels of interest using the t-test and a posterior reduction of the feature dimension using factor analysis. Factor loadings are used as features for three different classifiers: Two multivariate Gaussian mixture model, with linear and quadratic discriminant function, and a support vector machine with linear kernel.

RESULTS

An accuracy rate up to 95% when NC and AD are considered and an accuracy rate up to 88% and 86% for NC-MCI and NC-MCI,AD, respectively, are obtained using SVM with linear kernel.

CONCLUSIONS

Results are compared to the voxel-as-features and a PCA- based approach and the proposed methodology achieves better classification performance.

Computer-aided diagnosis of Alzheimer's disease using support vector machines and classification trees

This paper presents a computer-aided diagnosis technique for improving the accuracy of early diagnosis of Alzheimer-type dementia. The proposed methodology is based on the selection of voxels which present Welch's t-test between both classes, normal and Alzheimer images, greater than a given threshold. The mean and standard deviation of intensity values are calculated for selected voxels. They are chosen as feature vectors for two different classifiers: support vector machines with linear kernel and classification trees. The proposed methodology reaches greater than 95% accuracy in the classification task.

Classification of functional brain images using a GMM-based multi-variate approach

This paper presents a novel method for automatic selection of regions of interest (ROIs) of functional brain images based on Gaussian mixture models (GMM), which relieves the so-called small size sample problem in the classification of functional brain images for the diagnosis of Alzheimer's disease (AD). In a first step, brain images are preprocessed in order to find an average image including differences between controls and AD patients. Then, ROIs are extracted using a GMM which is adjusted by using the expectation maximization (EM) algorithm. This reduced set of features provides the activation map of each patient and allows us to train statistical classifiers based on support vector machines (SVMs). The leave-one-out cross-validation technique is used to validate the results obtained by the supervised learning-based computer aided diagnosis (CAD) system over databases of SPECT and PET images yielding an accuracy rate up to 96.67%.

SVM-based CAD system for early detection of the Alzheimer's disease using kernel PCA and LDA

Single-photon emission tomography (SPECT) imaging has been widely used to guide clinicians in the early Alzheimer's disease (AD) diagnosis challenge. However, AD detection still relies on subjective steps carried out by clinicians, which entail in some way subjectivity to the final diagnosis. In this work, kernel principal component analysis (PCA) and linear discriminant analysis (LDA) are applied on functional images as dimension reduction and feature extraction techniques, which are subsequently used to train a supervised support vector machine (SVM) classifier. The complete methodology provides a kernel-based computer-aided diagnosis (CAD) system capable to distinguish AD from normal subjects with 92.31% accuracy rate for a SPECT database consisting of 91 patients. The proposed methodology outperforms voxels-as-features (VAF) that was considered as baseline approach, which yields 80.22% for the same SPECT database.

Antiphospholipid syndrome: a comprehensive review of a complex and multisystemic disease

The antiphospholipid syndrome (APS) is an acquired thombophilia, which is characterized by one or more thrombotic episodes and obstetric complications in the presence of antiphospholipid (aPL) antibodies (Abs). APL Abs are detected by laboratory tests such as lupus anticoagulant (LAC), anticardiolipin (aCL) and anti-Beta2-glycoprotein I (Beta2GPI) Abs. This article reviews the most current aspects of APS with emphasis on the pathophysiology of the disease, clinical manifestations, laboratory tests, and current modalities of treatment.

SVM-based computer-aided diagnosis of the Alzheimer's disease using t-test NMSE feature selection with feature correlation weighting

This letter shows a computer-aided diagnosis (CAD) technique for the early detection of the Alzheimer's disease (AD) based on single photon emission computed tomography (SPECT) image feature selection and a statistical learning theory classifier. The challenge of the curse of dimensionality is addressed by reducing the large dimensionality of the input data and defining normalized mean squared error features over regions of interest (ROI) that are selected by a t-test feature selection with feature correlation weighting. Thus, normalized mean square error (NMSE) features of cubic blocks located in the temporo-parietal brain region yields peak accuracy values of 98.3% for almost linear kernel support vector machine (SVM) defined over the 20 most discriminative features extracted. This new method outperformed recent developed methods for early AD diagnosis.