Support vector machine-based classification of Alzheimer's disease from whole-brain anatomical MRI

Identification of Conversion from Normal Elderly Cognition to Alzheimer's Disease using Multimodal Support Vector Machine

Alzheimer's disease (AD) is one of the most serious progressive neurodegenerative diseases among the elderly, therefore the identification of conversion to AD at the earlier stage has become a crucial issue. In this study, we applied multimodal support vector machine to identify the conversion from normal elderly cognition to mild cognitive impairment (MCI) or AD based on magnetic resonance imaging and positron emission tomography data. The participants included two independent cohorts (Training set: 121 AD patients and 120 normal controls (NC); Testing set: 20 NC converters and 20 NC non-converters) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. The multimodal results showed that the accuracy, sensitivity, and specificity of the classification between NC converters and NC non-converters were 67.5% , 73.33% , and 64% , respectively. Furthermore, the classification results with feature selection increased to 70% accuracy, 75% sensitivity, and 66.67% specificity. The classification results using multimodal data are markedly superior to that using a single modality when we identified the conversion from NC to MCI or AD. The model built in this study of identifying the risk of normal elderly converting to MCI or AD will be helpful in clinical diagnosis and pathological research.

A review of feature reduction techniques in neuroimaging

Machine learning techniques are increasingly being used in making relevant predictions and inferences on individual subjects neuroimaging scan data. Previous studies have mostly focused on categorical discrimination of patients and matched healthy controls and more recently, on prediction of individual continuous variables such as clinical scores or age. However, these studies are greatly hampered by the large number of predictor variables (voxels) and low observations (subjects) also known as the curse-of-dimensionality or small-n-large-p problem. As a result, feature reduction techniques such as feature subset selection and dimensionality reduction are used to remove redundant predictor variables and experimental noise, a process which mitigates the curse-of-dimensionality and small-n-large-p effects. Feature reduction is an essential step before training a machine learning model to avoid overfitting and therefore improving model prediction accuracy and generalization ability. In this review, we discuss feature reduction techniques used with machine learning in neuroimaging studies.

Framingham Coronary Heart Disease Risk Score Can be Predicted from Structural Brain Images in Elderly Subjects

Recent literature has presented evidence that cardiovascular risk factors (CVRF) play an important role on cognitive performance in elderly individuals, both those who are asymptomatic and those who suffer from symptoms of neurodegenerative disorders. Findings from studies applying neuroimaging methods have increasingly reinforced such notion. Studies addressing the impact of CVRF on brain anatomy changes have gained increasing importance, as recent papers have reported gray matter loss predominantly in regions traditionally affected in Alzheimer's disease (AD) and vascular dementia in the presence of a high degree of cardiovascular risk. In the present paper, we explore the association between CVRF and brain changes using pattern recognition techniques applied to structural MRI and the Framingham score (a composite measure of cardiovascular risk largely used in epidemiological studies) in a sample of healthy elderly individuals. We aim to answer the following questions: is it possible to decode (i.e., to learn information regarding cardiovascular risk from structural brain images) enabling individual predictions? Among clinical measures comprising the Framingham score, are there particular risk factors that stand as more predictable from patterns of brain changes? Our main findings are threefold: (i) we verified that structural changes in spatially distributed patterns in the brain enable statistically significant prediction of Framingham scores. This result is still significant when controlling for the presence of the APOE 4 allele (an important genetic risk factor for both AD and cardiovascular disease). (ii) When considering each risk factor singly, we found different levels of correlation between real and predicted factors; however, single factors were not significantly predictable from brain images when considering APOE4 allele presence as covariate. (iii) We found important gender differences, and the possible causes of that finding are discussed.

Statistical sex determination from craniometrics: Comparison of linear discriminant analysis, logistic regression, and support vector machines

Accuracy of identification tools in forensic anthropology primarily rely upon the variations inherent in the data upon which they are built. Sex determination methods based on craniometrics are widely used and known to be specific to several factors (e.g. sample distribution, population, age, secular trends, measurement technique, etc.). The goal of this study is to discuss the potential variations linked to the statistical treatment of the data. Traditional craniometrics of four samples extracted from documented osteological collections (from Portugal, France, the U.S.A., and Thailand) were used to test three different classification methods: linear discriminant analysis (LDA), logistic regression (LR), and support vector machines (SVM). The Portuguese sample was set as a training model on which the other samples were applied in order to assess the validity and reliability of the different models. The tests were performed using different parameters: some included the selection of the best predictors; some included a strict decision threshold (sex assessed only if the related posterior probability was high, including the notion of indeterminate result); and some used an unbalanced sex-ratio. Results indicated that LR tends to perform slightly better than the other techniques and offers a better selection of predictors. Also, the use of a decision threshold (i.e. p>0.95) is essential to ensure an acceptable reliability of sex determination methods based on craniometrics. Although the Portuguese, French, and American samples share a similar sexual dimorphism, application of Western models on the Thai sample (that displayed a lower degree of dimorphism) was unsuccessful.

Integrated cortical structural marker for Alzheimer's disease

In this article, we propose an approach to integrate cortical morphology measures for improving the discrimination of individuals with and without very mild Alzheimer's disease (AD). FreeSurfer was applied to scans collected from 83 participants with very mild AD and 124 cognitively normal individuals. We generated cortex thickness, white matter convexity (aka "sulcal depth"), and white matter surface metric distortion measures on a normalized surface atlas in this first study to integrate high resolution gray matter thickness and white matter surface geometric measures in identifying very mild AD. Principal component analysis was applied to each individual structural measure to generate eigenvectors. Discrimination power based on individual and combined measures are compared, based on stepwise logistic regression and 10-fold cross-validation. Global AD likelihood index and surface-based likelihood maps were also generated. Our results show complementary patterns on the cortical surface between thickness, which reflects gray matter atrophy, convexity, which reflects white matter sulcal depth changes and metric distortion, which reflects white matter surface area changes. The classifier integrating all 3 types of surface measures significantly improved classification performance compared with classification based on single measures. The principal component analysis-based approach provides a framework for achieving high discrimination power by integrating high-dimensional data, and this method could be very powerful in future studies for early diagnosis of diseases that are known to be associated with abnormal gyral and sulcal patterns.

An ensemble-of-classifiers based approach for early diagnosis of Alzheimer's disease: classification using structural features of brain images

Structural brain imaging is playing a vital role in identification of changes that occur in brain associated with Alzheimer's disease. This paper proposes an automated image processing based approach for the identification of AD from MRI of the brain. The proposed approach is novel in a sense that it has higher specificity/accuracy values despite the use of smaller feature set as compared to existing approaches. Moreover, the proposed approach is capable of identifying AD patients in early stages. The dataset selected consists of 85 age and gender matched individuals from OASIS database. The features selected are volume of GM, WM, and CSF and size of hippocampus. Three different classification models (SVM, MLP, and J48) are used for identification of patients and controls. In addition, an ensemble of classifiers, based on majority voting, is adopted to overcome the error caused by an independent base classifier. Ten-fold cross validation strategy is applied for the evaluation of our scheme. Moreover, to evaluate the performance of proposed approach, individual features and combination of features are fed to individual classifiers and ensemble based classifier. Using size of left hippocampus as feature, the accuracy achieved with ensemble of classifiers is 93.75%, with 100% specificity and 87.5% sensitivity.

Diagnostic classification of arterial spin labeling and structural MRI in presenile early stage dementia

Because hypoperfusion of brain tissue precedes atrophy in dementia, the detection of dementia may be advanced by the use of perfusion information. Such information can be obtained noninvasively with arterial spin labeling (ASL), a relatively new MR technique quantifying cerebral blood flow (CBF). Using ASL and structural MRI, we evaluated diagnostic classification in 32 prospectively included presenile early stage dementia patients and 32 healthy controls. Patients were suspected of Alzheimer's disease (AD) or frontotemporal dementia. Classification was based on CBF as perfusion marker, gray matter (GM) volume as atrophy marker, and their combination. These markers were each examined using six feature extraction methods: a voxel-wise method and a region of interest (ROI)-wise approach using five ROI-sets in the GM. These ROI-sets ranged in number from 72 brain regions to a single ROI for the entire supratentorial brain. Classification was performed with a linear support vector machine classifier. For validation of the classification method on the basis of GM features, a reference dataset from the AD Neuroimaging Initiative database was used consisting of AD patients and healthy controls. In our early stage dementia population, the voxelwise feature-extraction approach achieved more accurate results (area under the curve (AUC) range = 86 - 91%) than all other approaches (AUC = 57 - 84%). Used in isolation, CBF quantified with ASL was a good diagnostic marker for dementia. However, our findings indicated only little added diagnostic value when combining ASL with the structural MRI data (AUC = 91%), which did not significantly improve over accuracy of structural MRI atrophy marker by itself.

Application of a MRI based index to longitudinal atrophy change in Alzheimer disease, mild cognitive impairment and healthy older individuals in the AddNeuroMed cohort

Cross sectional studies of patients at risk of developing Alzheimer disease (AD) have identified several brain regions known to be prone to degeneration suitable as biomarkers, including hippocampal, ventricular, and whole brain volume. The aim of this study was to longitudinally evaluate an index based on morphometric measures derived from MRI data that could be used for classification of AD and healthy control subjects, as well as prediction of conversion from mild cognitive impairment (MCI) to AD. Patients originated from the AddNeuroMed project at baseline (119 AD, 119 MCI, 110 controls (CTL)) and 1-year follow-up (62 AD, 73 MCI, 79 CTL). Data consisted of 3D T1-weighted MR images, demographics, MMSE, ADAS-Cog, CERAD and CDR scores, and APOE e4 status. We computed an index using a multivariate classification model (AD vs. CTL), using orthogonal partial least squares to latent structures (OPLS). Sensitivity, specificity and AUC were determined. Performance of the classifier (AD vs. CTL) was high at baseline (10-fold cross-validation, 84% sensitivity, 91% specificity, 0.93 AUC) and at 1-year follow-up (92% sensitivity, 74% specificity, 0.93 AUC). Predictions of conversion from MCI to AD were good at baseline (77% of MCI converters) and at follow-up (91% of MCI converters). MCI carriers of the APOE e4 allele manifested more atrophy and presented a faster cognitive decline when compared to non-carriers. The derived index demonstrated a steady increase in atrophy over time, yielding higher accuracy in prediction at the time of clinical conversion. Neuropsychological tests appeared less sensitive to changes over time. However, taking the average of the two time points yielded better correlation between the index and cognitive scores as opposed to using cross-sectional data only. Thus, multivariate classification seemed to detect patterns of AD changes before conversion from MCI to AD and including longitudinal information is of great importance.

Prediction of myelopathic level in cervical spondylotic myelopathy using diffusion tensor imaging

PURPOSE

To investigate the use of a newly designed machine learning-based classifier in the automatic identification of myelopathic levels in cervical spondylotic myelopathy (CSM).

MATERIALS AND METHODS

In all, 58 normal volunteers and 16 subjects with CSM were recruited for diffusion tensor imaging (DTI) acquisition. The eigenvalues were extracted as the selected features from DTI images. Three classifiers, naive Bayesian, support vector machine, and support tensor machine, and fractional anisotropy (FA) were employed to identify myelopathic levels. The results were compared with clinical level diagnosis results and accuracy, sensitivity, and specificity were calculated to evaluate the performance of the developed classifiers.

RESULTS

The accuracy by support tensor machine was the highest (93.62%) among the three classifiers. The support tensor machine also showed excellent capacity to identify true positives (sensitivity: 84.62%) and true negatives (specificity: 97.06%). The accuracy by FA value was the lowest (76%) in all the methods.

CONCLUSION

The classifiers-based method using eigenvalues had a better performance in identifying the levels of CSM than the diagnosis using FA values. The support tensor machine was the best among three classifiers.

Advancing research diagnostic criteria for Alzheimer's disease: the IWG-2 criteria

In the past 8 years, both the International Working Group (IWG) and the US National Institute on Aging-Alzheimer's Association have contributed criteria for the diagnosis of Alzheimer's disease (AD) that better define clinical phenotypes and integrate biomarkers into the diagnostic process, covering the full staging of the disease. This Position Paper considers the strengths and limitations of the IWG research diagnostic criteria and proposes advances to improve the diagnostic framework. On the basis of these refinements, the diagnosis of AD can be simplified, requiring the presence of an appropriate clinical AD phenotype (typical or atypical) and a pathophysiological biomarker consistent with the presence of Alzheimer's pathology. We propose that downstream topographical biomarkers of the disease, such as volumetric MRI and fluorodeoxyglucose PET, might better serve in the measurement and monitoring of the course of disease. This paper also elaborates on the specific diagnostic criteria for atypical forms of AD, for mixed AD, and for the preclinical states of AD.

Diagnosis of autism spectrum disorders using regional and interregional morphological features

This article describes a novel approach to identify autism spectrum disorder (ASD) utilizing regional and interregional morphological patterns extracted from structural magnetic resonance images. Two types of features are extracted to characterize the morphological patterns: (1) Regional features, which includes the cortical thickness, volumes of cortical gray matter, and cortical-associated white matter regions, and several subcortical structures extracted from different regions-of-interest (ROIs); (2) Interregional features, which convey the morphological change pattern between pairs of ROIs. We demonstrate that the integration of regional and interregional features via multi-kernel learning technique can significantly improve the classification performance of ASD, compared with using either regional or interregional features alone. Specifically, the proposed framework achieves an accuracy of 96.27% and an area of 0.9952 under the receiver operating characteristic curve, indicating excellent diagnostic power and generalizability. The best performance is achieved when both feature types are weighted approximately equal, indicating complementary between these two feature types. Regions that contributed the most to classification are in line with those reported in the previous studies, particularly the subcortical structures that are highly associated with human emotional modulation and memory formation. The autistic brains demonstrate a significant rightward asymmetry pattern particularly in the auditory language areas. These findings are in agreement with the fact that ASD is a behavioral- and language-related neurodevelopmental disorder. By concurrent consideration of both regional and interregional features, the current work presents an effective means for better characterization of neurobiological underpinnings of ASD that facilitates its identification from typically developing children.

Extracting salient brain patterns for imaging-based classification of neurodegenerative diseases

Neurodegenerative diseases comprise a wide variety of mental symptoms whose evolution is not directly related to the visual analysis made by radiologists, who can hardly quantify systematic differences. Moreover, automatic brain morphometric analyses, that do perform this quantification, contribute very little to the comprehension of the disease, i.e., many of these methods classify but they do not produce useful anatomo-functional correlations. This paper presents a new fully automatic image analysis method that reveals discriminative brain patterns associated to the presence of neurodegenerative diseases, mining systematic differences and therefore grading objectively any neurological disorder. This is accomplished by a fusion strategy that mixes together bottom-up and top-down information flows. Bottom-up information comes from a multiscale analysis of different image features, while the top-down stage includes learning and fusion strategies formulated as a max-margin multiple-kernel optimization problem. The capacity of finding discriminative anatomic patterns was evaluated using the Alzheimer's disease (AD) as the use case. The classification performance was assessed under different configurations of the proposed approach in two public brain magnetic resonance datasets (OASIS-MIRIAD) with patients diagnosed with AD, showing an improvement varying from 6.2% to 13% in the equal error rate measure, with respect to what has been reported by the feature-based morphometry strategy. In terms of the anatomical analysis, discriminant regions found by the proposed approach highly correlates to what has been reported in clinical studies of AD.

Multi-task linear programming discriminant analysis for the identification of progressive MCI individuals

Accurately identifying mild cognitive impairment (MCI) individuals who will progress to Alzheimer's disease (AD) is very important for making early interventions. Many classification methods focus on integrating multiple imaging modalities such as magnetic resonance imaging (MRI) and fluorodeoxyglucose positron emission tomography (FDG-PET). However, the main challenge for MCI classification using multiple imaging modalities is the existence of a lot of missing data in many subjects. For example, in the Alzheimer's Disease Neuroimaging Initiative (ADNI) study, almost half of the subjects do not have PET images. In this paper, we propose a new and flexible binary classification method, namely Multi-task Linear Programming Discriminant (MLPD) analysis, for the incomplete multi-source feature learning. Specifically, we decompose the classification problem into different classification tasks, i.e., one for each combination of available data sources. To solve all different classification tasks jointly, our proposed MLPD method links them together by constraining them to achieve the similar estimated mean difference between the two classes (under classification) for those shared features. Compared with the state-of-the-art incomplete Multi-Source Feature (iMSF) learning method, instead of constraining different classification tasks to choose a common feature subset for those shared features, MLPD can flexibly and adaptively choose different feature subsets for different classification tasks. Furthermore, our proposed MLPD method can be efficiently implemented by linear programming. To validate our MLPD method, we perform experiments on the ADNI baseline dataset with the incomplete MRI and PET images from 167 progressive MCI (pMCI) subjects and 226 stable MCI (sMCI) subjects. We further compared our method with the iMSF method (using incomplete MRI and PET images) and also the single-task classification method (using only MRI or only subjects with both MRI and PET images). Experimental results show very promising performance of our proposed MLPD method.

Automatic ROI selection in structural brain MRI using SOM 3D projection

This paper presents a method for selecting Regions of Interest (ROI) in brain Magnetic Resonance Imaging (MRI) for diagnostic purposes, using statistical learning and vector quantization techniques. The proposed method models the distribution of GM and WM tissues grouping the voxels belonging to each tissue in ROIs associated to a specific neurological disorder. Tissue distribution of normal and abnormal images is modelled by a Self-Organizing map (SOM), generating a set of representative prototypes, and the receptive field (RF) of each SOM prototype defines a ROI. Moreover, the proposed method computes the relative importance of each ROI by means of its discriminative power. The devised method has been assessed using 818 images from the Alzheimer's disease Neuroimaging Initiative (ADNI) which were previously segmented through Statistical Parametric Mapping (SPM). The proposed algorithm was used over these images to parcel ROIs associated to the Alzheimer's Disease (AD). Additionally, this method can be used to extract a reduced set of discriminative features for classification, since it compresses discriminative information contained in the brain. Voxels marked by ROIs which were computed using the proposed method, yield classification results up to 90% of accuracy for controls (CN) and Alzheimer's disease (AD) patients, and 84% of accuracy for Mild Cognitive Impairment (MCI) and AD patients.

Longitudinal deformation models, spatial regularizations and learning strategies to quantify Alzheimer's disease progression

In the context of Alzheimer's disease, two challenging issues are (1) the characterization of local hippocampal shape changes specific to disease progression and (2) the identification of mild-cognitive impairment patients likely to convert. In the literature, (1) is usually solved first to detect areas potentially related to the disease. These areas are then considered as an input to solve (2). As an alternative to this sequential strategy, we investigate the use of a classification model using logistic regression to address both issues (1) and (2) simultaneously. The classification of the patients therefore does not require any a priori definition of the most representative hippocampal areas potentially related to the disease, as they are automatically detected. We first quantify deformations of patients' hippocampi between two time points using the large deformations by diffeomorphisms framework and transport these deformations to a common template. Since the deformations are expected to be spatially structured, we perform classification combining logistic loss and spatial regularization techniques, which have not been explored so far in this context, as far as we know. The main contribution of this paper is the comparison of regularization techniques enforcing the coefficient maps to be spatially smooth (Sobolev), piecewise constant (total variation) or sparse (fused LASSO) with standard regularization techniques which do not take into account the spatial structure (LASSO, ridge and ElasticNet). On a dataset of 103 patients out of ADNI, the techniques using spatial regularizations lead to the best classification rates. They also find coherent areas related to the disease progression.

Individual classification of children with epilepsy using support vector machine with multiple indices of diffusion tensor imaging

INTRODUCTION

Support vector machines (SVM) have recently been demonstrated to be useful for voxel-based MR image classification. In the present study we sought to evaluate whether this method is feasible in the classification of childhood epilepsy intractability based on diffusion tensor imaging (DTI), with adequate accuracy. We applied SVM in conjunction DTI indices of fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD) and axial diffusivity (AD). DTI studies have reported white matter abnormalities in childhood-onset epilepsy, but the mechanisms underlying these abnormalities are not well understood. The aim of this study was to examine the relationship between epileptic seizures and cerebral white matter abnormalities identified by DTI in children with active compared to remitted epilepsy utilizing an automated and unsupervised classification method.

METHODS

The DTI data were tensor-derived indices including FA, MD, AD and RD in 49 participants including 20 children with epilepsy 5-6 years after seizure onset as compared to healthy controls. To determine whether there was normalization of white matter diffusion behavior following cessation of seizures and treatment, the epilepsy subjects were grouped into those with active versus remitted epilepsy. Group comparisons were previously made examining FA, MD and RD via whole-brain tract-based spatial statistics (TBSS). The SVM analysis was undertaken with the WEKA software package with 10-fold cross validation. Weighted sensitivity, specificity and accuracy were measured for all the DTI indices for two classifications: (1) controls vs. all children with epilepsy and (2) controls vs. children with remitted epilepsy vs. children with active epilepsy.

RESULTS

Using TBSS, significant differences were identified between controls and all children with epilepsy, between controls and children with active epilepsy, and also between the active and remitted epilepsy groups. There were no significant differences between the remitted epilepsy and controls on any DTI measure. In the SVM analysis, the best predictor between controls and all children with epilepsy was MD, with a sensitivity of 90-100% and a specificity between 96.6 and 100%. For the three-way classification, the best results were for FA with 100% sensitivity and specificity.

CONCLUSION

DTI-based SVM classification appears promising for distinguishing children with active epilepsy from either those with remitted epilepsy or controls, and the question that arises is whether it will prove useful as a prognostic index of seizure remission. While SVM can correctly identify children with active epilepsy from other groups' diagnosis, further research is needed to determine the efficacy of SVM as a prognostic tool in longitudinal clinical studies.

Development of a brain MRI-based hidden Markov model for dementia recognition

BACKGROUND

Dementia is an age-related cognitive decline which is indicated by an early degeneration of cortical and sub-cortical structures. Characterizing those morphological changes can help to understand the disease development and contribute to disease early prediction and prevention. But modeling that can best capture brain structural variability and can be valid in both disease classification and interpretation is extremely challenging. The current study aimed to establish a computational approach for modeling the magnetic resonance imaging (MRI)-based structural complexity of the brain using the framework of hidden Markov models (HMMs) for dementia recognition.

METHODS

Regularity dimension and semi-variogram were used to extract structural features of the brains, and vector quantization method was applied to convert extracted feature vectors to prototype vectors. The output VQ indices were then utilized to estimate parameters for HMMs. To validate its accuracy and robustness, experiments were carried out on individuals who were characterized as non-demented and mild Alzheimer's diseased. Four HMMs were constructed based on the cohort of non-demented young, middle-aged, elder and demented elder subjects separately. Classification was carried out using a data set including both non-demented and demented individuals with a wide age range.

RESULTS

The proposed HMMs have succeeded in recognition of individual who has mild Alzheimer's disease and achieved a better classification accuracy compared to other related works using different classifiers. Results have shown the ability of the proposed modeling for recognition of early dementia.

CONCLUSION

The findings from this research will allow individual classification to support the early diagnosis and prediction of dementia. By using the brain MRI-based HMMs developed in our proposed research, it will be more efficient, robust and can be easily used by clinicians as a computer-aid tool for validating imaging bio-markers for early prediction of dementia.

Regions of interest computed by SVM wrapped method for Alzheimer's disease examination from segmented MRI

Accurate identification of the most relevant brain regions linked to Alzheimer's disease (AD) is crucial in order to improve diagnosis techniques and to better understand this neurodegenerative process. For this purpose, statistical classification is suitable. In this work, a novel method based on support vector machine recursive feature elimination (SVM-RFE) is proposed to be applied on segmented brain MRI for detecting the most discriminant AD regions of interest (ROIs). The analyses are performed both on gray and white matter tissues, achieving up to 100% accuracy after classification and outperforming the results obtained by the standard t-test feature selection. The present method, applied on different subject sets, permits automatically determining high-resolution areas surrounding the hippocampal area without needing to divide the brain images according to any common template.

Aberrant functional connectivity for diagnosis of major depressive disorder: a discriminant analysis

AIM

Aberrant brain functional connectivity patterns have been reported in major depressive disorder (MDD). It is unknown whether they can be used in discriminant analysis for diagnosis of MDD. In the present study we examined the efficiency of discriminant analysis of MDD by individualized computer-assisted diagnosis.

METHODS

Based on resting-state functional magnetic resonance imaging data, a new approach was adopted to investigate functional connectivity changes in 39 MDD patients and 37 well-matched healthy controls. By using the proposed feature selection method, we identified significant altered functional connections in patients. They were subsequently applied to our analysis as discriminant features using a support vector machine classification method. Furthermore, the relative contribution of functional connectivity was estimated.

RESULTS

After subset selection of high-dimension features, the support vector machine classifier reached up to approximately 84% with leave-one-out training during the discrimination process. Through summarizing the classification contribution of functional connectivities, we obtained four obvious contribution modules: inferior orbitofrontal module, supramarginal gyrus module, inferior parietal lobule-posterior cingulated gyrus module and middle temporal gyrus-inferior temporal gyrus module.

CONCLUSION

The experimental results demonstrated that the proposed method is effective in discriminating MDD patients from healthy controls. Functional connectivities might be useful as new biomarkers to assist clinicians in computer auxiliary diagnosis of MDD.

Fast parallel image registration on CPU and GPU for diagnostic classification of Alzheimer's disease

Nonrigid image registration is an important, but time-consuming task in medical image analysis. In typical neuroimaging studies, multiple image registrations are performed, i.e., for atlas-based segmentation or template construction. Faster image registration routines would therefore be beneficial. In this paper we explore acceleration of the image registration package elastix by a combination of several techniques: (i) parallelization on the CPU, to speed up the cost function derivative calculation; (ii) parallelization on the GPU building on and extending the OpenCL framework from ITKv4, to speed up the Gaussian pyramid computation and the image resampling step; (iii) exploitation of certain properties of the B-spline transformation model; (iv) further software optimizations. The accelerated registration tool is employed in a study on diagnostic classification of Alzheimer's disease and cognitively normal controls based on T1-weighted MRI. We selected 299 participants from the publicly available Alzheimer's Disease Neuroimaging Initiative database. Classification is performed with a support vector machine based on gray matter volumes as a marker for atrophy. We evaluated two types of strategies (voxel-wise and region-wise) that heavily rely on nonrigid image registration. Parallelization and optimization resulted in an acceleration factor of 4-5x on an 8-core machine. Using OpenCL a speedup factor of 2 was realized for computation of the Gaussian pyramids, and 15-60 for the resampling step, for larger images. The voxel-wise and the region-wise classification methods had an area under the receiver operator characteristic curve of 88 and 90%, respectively, both for standard and accelerated registration. We conclude that the image registration package elastix was substantially accelerated, with nearly identical results to the non-optimized version. The new functionality will become available in the next release of elastix as open source under the BSD license.

Improved volumetric measurement of brain structure with a distortion correction procedure using an ADNI phantom

PURPOSE

Serial magnetic resonance imaging (MRI) images acquired from multisite and multivendor MRI scanners are widely used in measuring longitudinal structural changes in the brain. Precise and accurate measurements are important in understanding the natural progression of neurodegenerative disorders such as Alzheimer's disease. However, geometric distortions in MRI images decrease the accuracy and precision of volumetric or morphometric measurements. To solve this problem, the authors suggest a commercially available phantom-based distortion correction method that accommodates the variation in geometric distortion within MRI images obtained with multivendor MRI scanners.

METHODS

The authors' method is based on image warping using a polynomial function. The method detects fiducial points within a phantom image using phantom analysis software developed by the Mayo Clinic and calculates warping functions for distortion correction. To quantify the effectiveness of the authors' method, the authors corrected phantom images obtained from multivendor MRI scanners and calculated the root-mean-square (RMS) of fiducial errors and the circularity ratio as evaluation values. The authors also compared the performance of the authors' method with that of a distortion correction method based on a spherical harmonics description of the generic gradient design parameters. Moreover, the authors evaluated whether this correction improves the test-retest reproducibility of voxel-based morphometry in human studies.

RESULTS

A Wilcoxon signed-rank test with uncorrected and corrected images was performed. The root-mean-square errors and circularity ratios for all slices significantly improved (p < 0.0001) after the authors' distortion correction. Additionally, the authors' method was significantly better than a distortion correction method based on a description of spherical harmonics in improving the distortion of root-mean-square errors (p < 0.001 and 0.0337, respectively). Moreover, the authors' method reduced the RMS error arising from gradient nonlinearity more than gradwarp methods. In human studies, the coefficient of variation of voxel-based morphometry analysis of the whole brain improved significantly from 3.46% to 2.70% after distortion correction of the whole gray matter using the authors' method (Wilcoxon signed-rank test, p < 0.05).

CONCLUSIONS

The authors proposed a phantom-based distortion correction method to improve reproducibility in longitudinal structural brain analysis using multivendor MRI. The authors evaluated the authors' method for phantom images in terms of two geometrical values and for human images in terms of test-retest reproducibility. The results showed that distortion was corrected significantly using the authors' method. In human studies, the reproducibility of voxel-based morphometry analysis for the whole gray matter significantly improved after distortion correction using the authors' method.

Significance of normalization on anatomical MRI measures in predicting Alzheimer's disease

This study establishes a new approach for combining neuroimaging and neuropsychological measures for an optimal decisional space to classify subjects with Alzheimer's disease (AD). This approach relies on a multivariate feature selection method with different MRI normalization techniques. Subcortical volume, cortical thickness, and surface area measures are obtained using MRIs from 189 participants (129 normal controls and 60 AD patients). Statistically significant variables were selected for each combination model to construct a multidimensional space for classification. Different normalization approaches were explored to gauge the effect on classification performance using a support vector machine classifier. Results indicate that the Mini-mental state examination (MMSE) measure is most discriminative among single-measure models, while subcortical volume combined with MMSE is the most effective multivariate model for AD classification. The study demonstrates that subcortical volumes need not be normalized, whereas cortical thickness should be normalized either by intracranial volume or mean thickness, and surface area is a weak indicator of AD with and without normalization. On the significant brain regions, a nearly perfect symmetry is observed for subcortical volumes and cortical thickness, and a significant reduction in thickness is particularly seen in the temporal lobe, which is associated with brain deficits characterizing AD.

Machine learning on brain MRI data for differential diagnosis of Parkinson's disease and Progressive Supranuclear Palsy

BACKGROUND

Supervised machine learning has been proposed as a revolutionary approach for identifying sensitive medical image biomarkers (or combination of them) allowing for automatic diagnosis of individual subjects. The aim of this work was to assess the feasibility of a supervised machine learning algorithm for the assisted diagnosis of patients with clinically diagnosed Parkinson's disease (PD) and Progressive Supranuclear Palsy (PSP).

METHOD

Morphological T1-weighted Magnetic Resonance Images (MRIs) of PD patients (28), PSP patients (28) and healthy control subjects (28) were used by a supervised machine learning algorithm based on the combination of Principal Components Analysis as feature extraction technique and on Support Vector Machines as classification algorithm. The algorithm was able to obtain voxel-based morphological biomarkers of PD and PSP.

RESULTS

The algorithm allowed individual diagnosis of PD versus controls, PSP versus controls and PSP versus PD with an Accuracy, Specificity and Sensitivity>90%. Voxels influencing classification between PD and PSP patients involved midbrain, pons, corpus callosum and thalamus, four critical regions known to be strongly involved in the pathophysiological mechanisms of PSP.

COMPARISON WITH EXISTING METHODS

Classification accuracy of individual PSP patients was consistent with previous manual morphological metrics and with other supervised machine learning application to MRI data, whereas accuracy in the detection of individual PD patients was significantly higher with our classification method.

CONCLUSIONS

The algorithm provides excellent discrimination of PD patients from PSP patients at an individual level, thus encouraging the application of computer-based diagnosis in clinical practice.

Extreme learning machine-based classification of ADHD using brain structural MRI data

BACKGROUND

Effective and accurate diagnosis of attention-deficit/hyperactivity disorder (ADHD) is currently of significant interest. ADHD has been associated with multiple cortical features from structural MRI data. However, most existing learning algorithms for ADHD identification contain obvious defects, such as time-consuming training, parameters selection, etc. The aims of this study were as follows: (1) Propose an ADHD classification model using the extreme learning machine (ELM) algorithm for automatic, efficient and objective clinical ADHD diagnosis. (2) Assess the computational efficiency and the effect of sample size on both ELM and support vector machine (SVM) methods and analyze which brain segments are involved in ADHD.

METHODS

High-resolution three-dimensional MR images were acquired from 55 ADHD subjects and 55 healthy controls. Multiple brain measures (cortical thickness, etc.) were calculated using a fully automated procedure in the FreeSurfer software package. In total, 340 cortical features were automatically extracted from 68 brain segments with 5 basic cortical features. F-score and SFS methods were adopted to select the optimal features for ADHD classification. Both ELM and SVM were evaluated for classification accuracy using leave-one-out cross-validation.

RESULTS

We achieved ADHD prediction accuracies of 90.18% for ELM using eleven combined features, 84.73% for SVM-Linear and 86.55% for SVM-RBF. Our results show that ELM has better computational efficiency and is more robust as sample size changes than is SVM for ADHD classification. The most pronounced differences between ADHD and healthy subjects were observed in the frontal lobe, temporal lobe, occipital lobe and insular.

CONCLUSION

Our ELM-based algorithm for ADHD diagnosis performs considerably better than the traditional SVM algorithm. This result suggests that ELM may be used for the clinical diagnosis of ADHD and the investigation of different brain diseases.

Alzheimer's disease detection in brain magnetic resonance images using multiscale fractal analysis

We present a new automated system for the detection of brain magnetic resonance images (MRI) affected by Alzheimer's disease (AD). The MRI is analyzed by means of multiscale analysis (MSA) to obtain its fractals at six different scales. The extracted fractals are used as features to differentiate healthy brain MRI from those of AD by a support vector machine (SVM) classifier. The result of classifying 93 brain MRIs consisting of 51 images of healthy brains and 42 of brains affected by AD, using leave-one-out cross-validation method, yielded 99.18% ± 0.01 classification accuracy, 100% sensitivity, and 98.20% ± 0.02 specificity. These results and a processing time of 5.64 seconds indicate that the proposed approach may be an efficient diagnostic aid for radiologists in the screening for AD.

Identification of mild Alzheimer's disease through automated classification of structural MRI features

The significant potential for early and accurate detection of Alzheimer's disease (AD) through neuroimaging data is becoming increasingly attractive in view of the possible advent of drugs which are able to modify or delay disease progression. In this paper, we aimed at developing an effective machine learning scheme which leverages structural magnetic resonance imaging features in order to identify and discriminate individuals affected by mild AD on a single subject basis. Selected features included one- and two-way combinations of subcortical and cortical volumes as well as cortical thickness and curvature of numerous brain regions which are known to be vulnerable to AD. Additionally, several feature combinations were fed into support vector machines (SVMs) as well as Naïve Bayes classifiers in order to compare scheme accuracy. The most efficient combination of features and classification scheme, which employed both subcortical and cortical volumes feature vectors and a SVM classifier, was able to distinguish mild AD patients from healthy controls with 86% accuracy (82% sensitivity and 90% specificity). While this investigation is of preliminary nature, and further efforts are currently underway towards automated feature selection, best classifier determination and parameter optimization, our results appear very promising in terms of automated high-accuracy discrimination of disease stages which cannot easily be distinguished though routine clinical investigation.

The utility of independent component analysis and machine learning in the identification of the amyotrophic lateral sclerosis diseased brain

Amyotrophic lateral sclerosis (ALS) is a devastating disease with a lifetime risk of ∼1 in 2000. Presently, diagnosis of ALS relies on clinical assessments for upper motor neuron and lower motor neuron deficits in multiple body segments together with a history of progression of symptoms. In addition, it is common to evaluate lower motor neuron pathology in ALS by electromyography. However, upper motor neuron pathology is solely assessed on clinical grounds, thus hindering diagnosis. In the past decade magnetic resonance methods have been shown to be sensitive to the ALS disease process, namely: resting-state connectivity measured with functional MRI, cortical thickness measured by high-resolution imaging, diffusion tensor imaging (DTI) metrics such as fractional anisotropy and radial diffusivity, and more recently magnetic resonance spectroscopy (MRS) measures of gamma-aminobutyric acid concentration. In this present work we utilize independent component analysis to derive brain networks based on resting-state functional magnetic resonance imaging and use those derived networks to build a disease state classifier using machine learning (support-vector machine). We show that it is possible to achieve over 71% accuracy for disease state classification. These results are promising for the development of a clinically relevant disease state classifier. Future inclusion of other MR modalities such as high-resolution structural imaging, DTI and MRS should improve this overall accuracy.

Different multivariate techniques for automated classification of MRI data in Alzheimer's disease and mild cognitive impairment

Automated structural magnetic resonance imaging (MRI) processing pipelines and different multivariate techniques are gaining popularity for Alzheimer's disease (AD) research. We used four supervised learning methods to classify AD patients and controls (CTL) and to prospectively predict the conversion of mild cognitive impairment (MCI) to AD from baseline MRI data. A total of 345 participants from the AddNeuroMed cohort were included in this study; 116 AD patients, 119 MCI patients and 110 CTL individuals. High resolution sagittal 3D MP-RAGE datasets were acquired and MRI data were processed using FreeSurfer. We explored the classification ability of orthogonal projections to latent structures (OPLS), decision trees (Trees), artificial neural networks (ANN) and support vector machines (SVM). Applying 10-fold cross-validation demonstrated that SVM and OPLS were slightly superior to Trees and ANN, although not statistically significant for distinguishing between AD and CTL. The classification experiments resulted in up to 83% sensitivity and 87% specificity for the best techniques. For the prediction of conversion of MCI patients at baseline to AD at 1-year follow-up, we obtained an accuracy of up to 86%. The value of the multivariate models derived from the classification of AD vs. CTL was shown to be robust and efficient in the identification of MCI converters.

Ensemble learning incorporating uncertain registration

This paper proposes a novel approach for improving the accuracy of statistical prediction methods in spatially normalized analysis. This is achieved by incorporating registration uncertainty into an ensemble learning scheme. A probabilistic registration method is used to estimate a distribution of probable mappings between subject and atlas space. This allows the estimation of the distribution of spatially normalized feature data, e.g., grey matter probability maps. From this distribution, samples are drawn for use as training examples. This allows the creation of multiple predictors, which are subsequently combined using an ensemble learning approach. Furthermore, extra testing samples can be generated to measure the uncertainty of prediction. This is applied to separating subjects with Alzheimer's disease from normal controls using a linear support vector machine on a region of interest in magnetic resonance images of the brain. We show that our proposed method leads to an improvement in discrimination using voxel-based morphometry and deformation tensor-based morphometry over bootstrap aggregating, a common ensemble learning framework. The proposed approach also generates more reasonable soft-classification predictions than bootstrap aggregating. We expect that this approach could be applied to other statistical prediction tasks where registration is important.

Hierarchical fusion of features and classifier decisions for Alzheimer's disease diagnosis

Pattern classification methods have been widely investigated for analysis of brain images to assist the diagnosis of Alzheimer's disease (AD) and its early stage such as mild cognitive impairment (MCI). By considering the nature of pathological changes, a large number of features related to both local brain regions and interbrain regions can be extracted for classification. However, it is challenging to design a single global classifier to integrate all these features for effective classification, due to the issue of small sample size. To this end, we propose a hierarchical ensemble classification method to combine multilevel classifiers by gradually integrating a large number of features from both local brain regions and interbrain regions. Thus, the large-scale classification problem can be divided into a set of small-scale and easier-to-solve problems in a bottom-up and local-to-global fashion, for more accurate classification. To demonstrate its performance, we use the spatially normalized grey matter (GM) of each MR brain image as imaging features. Specifically, we first partition the whole brain image into a number of local brain regions and, for each brain region, we build two low-level classifiers to transform local imaging features and the inter-region correlations into high-level features. Then, we generate multiple high-level classifiers, with each evaluating the high-level features from the respective brain regions. Finally, we combine the outputs of all high-level classifiers for making a final classification. Our method has been evaluated using the baseline MR images of 652 subjects (including 198 AD patients, 225 MCI patients, and 229 normal controls (NC)) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. The experimental results show that our classification method can achieve the accuracies of 92.0% and 85.3% for classifications of AD versus NC and MCI versus NC, respectively, demonstrating very promising classification performance compared to the state-of-the-art classification methods.

Sequential projection-based metacognitive learning in a radial basis function network for classification problems

In this paper, we present a sequential projection-based metacognitive learning algorithm in a radial basis function network (PBL-McRBFN) for classification problems. The algorithm is inspired by human metacognitive learning principles and has two components: a cognitive component and a metacognitive component. The cognitive component is a single-hidden-layer radial basis function network with evolving architecture. The metacognitive component controls the learning process in the cognitive component by choosing the best learning strategy for the current sample and adapts the learning strategies by implementing self-regulation. In addition, sample overlapping conditions and past knowledge of the samples in the form of pseudosamples are used for proper initialization of new hidden neurons to minimize the misclassification. The parameter update strategy uses projection-based direct minimization of hinge loss error. The interaction of the cognitive component and the metacognitive component addresses the what-to-learn, when-to-learn, and how-to-learn human learning principles efficiently. The performance of the PBL-McRBFN is evaluated using a set of benchmark classification problems from the University of California Irvine machine learning repository. The statistical performance evaluation on these problems proves the superior performance of the PBL-McRBFN classifier over results reported in the literature. Also, we evaluate the performance of the proposed algorithm on a practical Alzheimer's disease detection problem. The performance results on open access series of imaging studies and Alzheimer's disease neuroimaging initiative datasets, which are obtained from different demographic regions, clearly show that PBL-McRBFN can handle a problem with change in distribution.

A Decision Support System for the assisted diagnosis of brain tumors: a feasibility study for ¹⁸F-FDG PET preclinical studies

Decision support systems for the assisted medical diagnosis offer the main feature of giving assessments which are poorly affected from arbitrary clinical reasoning. Aim of this work was to assess the feasibility of a decision support system for the assisted diagnosis of brain cancer, such approach presenting potential for early diagnosis of tumors and for the classification of the degree of the disease progression. For this purpose, a supervised learning algorithm combined with a pattern recognition method was developed and cross-validated in ¹⁸F-FDG PET studies of a model of a brain tumour implantation.

Multivariate classification of patients with Alzheimer's and dementia with Lewy bodies using high-dimensional cortical thickness measurements: an MRI surface-based morphometric study

CONTEXT

Alzheimer's disease (AD) and dementia with Lewy bodies (DLB) are the most common neurodegenerative dementia types. It is important to differentiate between them because of the differences in prognosis and treatment approaches.

OBJECTIVE

Investigate if sparse partial least squares (SPLS) classification of cortical thickness measurements could differentiate between AD and DLB.

METHODS

Two independent cohorts without MR-protocol alignment in Norway and Slovenia with 97 AD and DLB subjects were enrolled. Cortical thickness measurements acquired with Freesurfer were used in subsequent SPLS classification runs. The cohorts were analyzed separately and afterwards combined. The models were trained with leave-one-out cross validation and test datasets where used when available. To study the impact of MR-protocol alignment, the classifiers were additionally tested on sets drawn exclusively from the independent cohorts.

RESULTS

The obtained sensitivity/specificity/AUC values were 94.4/88.89/0.978 and 88.2/94.1/0.969 in the Norwegian and Slovenian cohorts, respectively. Both cohorts showed AD-associated pattern of thinning in mid-anterior temporal, occipital and subgenual cingulate cortex, whereas the pattern supportive for DLB included thinning in dorsal cingulate, posterior temporal and lateral orbitofrontal regions. When combining the cohorts, sensitivity/specificity/AUC were 82.1/85.7/0.948 for the training and 77.8/75/0.731 for the testing datasets with the same pattern-of-difference. The models tested on datasets drawn exclusively from the independent cohorts did not produce adequate accuracy.

CONCLUSION

SPLS classification of cortical thickness is a good method for differentiating between AD and DLB, relatively stable even for mixed data, but not when tested on completely independent data drawn from different cohorts (without MR-protocol alignment).

The relevance voxel machine (RVoxM): a self-tuning Bayesian model for informative image-based prediction

This paper presents the relevance voxel machine (RVoxM), a dedicated Bayesian model for making predictions based on medical imaging data. In contrast to the generic machine learning algorithms that have often been used for this purpose, the method is designed to utilize a small number of spatially clustered sets of voxels that are particularly suited for clinical interpretation. RVoxM automatically tunes all its free parameters during the training phase, and offers the additional advantage of producing probabilistic prediction outcomes. We demonstrate RVoxM as a regression model by predicting age from volumetric gray matter segmentations, and as a classification model by distinguishing patients with Alzheimer's disease from healthy controls using surface-based cortical thickness data. Our results indicate that RVoxM yields biologically meaningful models, while providing state-of-the-art predictive accuracy.

Classification of structural MRI images in Alzheimer's disease from the perspective of ill-posed problems

BACKGROUND

Machine learning neuroimaging researchers have often relied on regularization techniques when classifying MRI images. Although these were originally introduced to deal with "ill-posed" problems it is rare to find studies that evaluate the ill-posedness of MRI image classification problems. In addition, to avoid the effects of the "curse of dimensionality" very often dimension reduction is applied to the data.

METHODOLOGY

Baseline structural MRI data from cognitively normal and Alzheimer's disease (AD) patients from the AD Neuroimaging Initiative database were used in this study. We evaluated here the ill-posedness of this classification problem across different dimensions and sample sizes and its relationship to the performance of regularized logistic regression (RLR), linear support vector machine (SVM) and linear regression classifier (LRC). In addition, these methods were compared with their principal components space counterparts.

PRINCIPAL FINDINGS

In voxel space the prediction performance of all methods increased as sample sizes increased. They were not only relatively robust to the increase of dimension, but they often showed improvements in accuracy. We linked this behavior to improvements in conditioning of the linear kernels matrices. In general the RLR and SVM performed similarly. Surprisingly, the LRC was often very competitive when the linear kernel matrices were best conditioned. Finally, when comparing these methods in voxel and principal component spaces, we did not find large differences in prediction performance.

CONCLUSIONS AND SIGNIFICANCE

We analyzed the problem of classifying AD MRI images from the perspective of linear ill-posed problems. We demonstrate empirically the impact of the linear kernel matrix conditioning on different classifiers' performance. This dependence is characterized across sample sizes and dimensions. In this context we also show that increased dimensionality does not necessarily degrade performance of machine learning methods. In general, this depends on the nature of the problem and the type of machine learning method.

Individual classification of mild cognitive impairment subtypes by support vector machine analysis of white matter DTI

BACKGROUND AND PURPOSE

MCI was recently subdivided into sd-aMCI, sd-fMCI, and md-aMCI. The current investigation aimed to discriminate between MCI subtypes by using DTI.

MATERIALS AND METHODS

Sixty-six prospective participants were included: 18 with sd-aMCI, 13 with sd-fMCI, and 35 with md-aMCI. Statistics included group comparisons using TBSS and individual classification using SVMs.

RESULTS

The group-level analysis revealed a decrease in FA in md-aMCI versus sd-aMCI in an extensive bilateral, right-dominant network, and a more pronounced reduction of FA in md-aMCI compared with sd-fMCI in right inferior fronto-occipital fasciculus and inferior longitudinal fasciculus. The comparison between sd-fMCI and sd-aMCI, as well as the analysis of the other diffusion parameters, yielded no significant group differences. The individual-level SVM analysis provided discrimination between the MCI subtypes with accuracies around 97%. The major limitation is the relatively small number of cases of MCI.

CONCLUSIONS

Our data show that, at the group level, the md-aMCI subgroup has the most pronounced damage in white matter integrity. Individually, SVM analysis of white matter FA provided highly accurate classification of MCI subtypes.

A Support Vector Machine based method to distinguish proteobacterial proteins from eukaryotic plant proteins

Background

Members of the phylum Proteobacteria are most prominent among bacteria causing plant diseases that result in a diminution of the quantity and quality of food produced by agriculture. To ameliorate these losses, there is a need to identify infections in early stages. Recent developments in next generation nucleic acid sequencing and mass spectrometry open the door to screening plants by the sequences of their macromolecules. Such an approach requires the ability to recognize the organismal origin of unknown DNA or peptide fragments. There are many ways to approach this problem but none have emerged as the best protocol. Here we attempt a systematic way to determine organismal origins of peptides by using a machine learning algorithm. The algorithm that we implement is a Support Vector Machine (SVM).

Result

The amino acid compositions of proteobacterial proteins were found to be different from those of plant proteins. We developed an SVM model based on amino acid and dipeptide compositions to distinguish between a proteobacterial protein and a plant protein. The amino acid composition (AAC) based SVM model had an accuracy of 92.44% with 0.85 Matthews correlation coefficient (MCC) while the dipeptide composition (DC) based SVM model had a maximum accuracy of 94.67% and 0.89 MCC. We also developed SVM models based on a hybrid approach (AAC and DC), which gave a maximum accuracy 94.86% and a 0.90 MCC. The models were tested on unseen or untrained datasets to assess their validity.

Conclusion

The results indicate that the SVM based on the AAC and DC hybrid approach can be used to distinguish proteobacterial from plant protein sequences.

Prediction of Alzheimer's disease and mild cognitive impairment using cortical morphological patterns

This article describes a novel approach to extract cortical morphological abnormality patterns from structural magnetic resonance imaging (MRI) data to improve the prediction accuracy of Alzheimer's disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI). Conventional approaches extract cortical morphological information, such as regional mean cortical thickness and regional cortical volumes, independently at different regions of interest (ROIs) without considering the relationship between these regions. Our approach involves constructing a similarity map where every element in the map represents the correlation of regional mean cortical thickness between a pair of ROIs. We will demonstrate in this article that this correlative morphological information gives significant improvement in classification performance when compared with ROI-based morphological information. Classification performance is further improved by integrating the correlative information with ROI-based information via multi-kernel support vector machines. This integrated framework achieves an accuracy of 92.35% for AD classification with an area of 0.9744 under the receiver operating characteristic (ROC) curve, and an accuracy of 83.75% for MCI classification with an area of 0.9233. In differentiating MCI subjects who converted to AD within 36 months from non-converters, an accuracy of 75.05% with an area of 0.8426 under ROC curve was achieved, indicating excellent diagnostic power and generalizability. The current work provides an alternative approach to extraction of high-order cortical information from structural MRI data for prediction of neurodegenerative diseases such as AD.

Multi-centre diagnostic classification of individual structural neuroimaging scans from patients with major depressive disorder

Quantitative abnormalities of brain structure in patients with major depressive disorder have been reported at a group level for decades. However, these structural differences appear subtle in comparison with conventional radiologically defined abnormalities, with considerable inter-subject variability. Consequently, it has not been possible to readily identify scans from patients with major depressive disorder at an individual level. Recently, machine learning techniques such as relevance vector machines and support vector machines have been applied to predictive classification of individual scans with variable success. Here we describe a novel hybrid method, which combines machine learning with feature selection and characterization, with the latter aimed at maximizing the accuracy of machine learning prediction. The method was tested using a multi-centre dataset of T(1)-weighted 'structural' scans. A total of 62 patients with major depressive disorder and matched controls were recruited from referred secondary care clinical populations in Aberdeen and Edinburgh, UK. The generalization ability and predictive accuracy of the classifiers was tested using data left out of the training process. High prediction accuracy was achieved (~90%). While feature selection was important for maximizing high predictive accuracy with machine learning, feature characterization contributed only a modest improvement to relevance vector machine-based prediction (~5%). Notably, while the only information provided for training the classifiers was T(1)-weighted scans plus a categorical label (major depressive disorder versus controls), both relevance vector machine and support vector machine 'weighting factors' (used for making predictions) correlated strongly with subjective ratings of illness severity. These results indicate that machine learning techniques have the potential to inform clinical practice and research, as they can make accurate predictions about brain scan data from individual subjects. Furthermore, machine learning weighting factors may reflect an objective biomarker of major depressive disorder illness severity, based on abnormalities of brain structure.

Comparison of four shape features for detecting hippocampal shape changes in early Alzheimer's

We compare four methods for generating shape-based features from 3D binary images of the hippocampus for use in group discrimination and classification. The first method we investigate is based on decomposing the hippocampal binary segmentation onto an orthonormal basis of spherical harmonics, followed by computation of shape invariants by tensor contraction using the Clebsch–Gordan coefficients. The second method we investigate is based on the classical 3D moment invariants; these are a special case of the spherical harmonics-based tensor invariants. The third method is based on solving the Helmholtz equation on the geometry of the binary hippocampal segmentation, and construction of shape-descriptive features from the eigenvalues of the Fourier-like modes of the geometry represented by the Laplacian eigenfunctions. The fourth method investigates the use of initial momentum obtained from the large-deformation diffeomorphic metric mapping method as a shape feature. Each of these shape features is tested for group differences in the control (Clinical Dementia Rating Scale CDR 0) and the early (very mild) Alzheimer's (CDR 0.5) population. Classification of individual shapes is performed via a linear support vector machine based classifer with leave-one-out cross validation to test for overall performance. These experiments show that all of these feature computation approaches gave stable and reasonable classification results on the same database, and with the same classifier. The best performance was achieved with the shape-features constructed from large-deformation diffeomorphic metric mapping-based initial momentum.

Pattern classification of large-scale functional brain networks: identification of informative neuroimaging markers for epilepsy

The accurate prediction of general neuropsychiatric disorders, on an individual basis, using resting-state functional magnetic resonance imaging (fMRI) is a challenging task of great clinical significance. Despite the progress to chart the differences between the healthy controls and patients at the group level, the pattern classification of functional brain networks across individuals is still less developed. In this paper we identify two novel neuroimaging measures that prove to be strongly predictive neuroimaging markers in pattern classification between healthy controls and general epileptic patients. These measures characterize two important aspects of the functional brain network in a quantitative manner: (i) coordinated operation among spatially distributed brain regions, and (ii) the asymmetry of bilaterally homologous brain regions, in terms of their global patterns of functional connectivity. This second measure offers a unique understanding of brain asymmetry at the network level, and, to the best of our knowledge, has not been previously used in pattern classification of functional brain networks. Using modern pattern-recognition approaches like sparse regression and support vector machine, we have achieved a cross-validated classification accuracy of 83.9% (specificity: 82.5%; sensitivity: 85%) across individuals from a large dataset consisting of 180 healthy controls and epileptic patients. We identified significantly changed functional pathways and subnetworks in epileptic patients that underlie the pathophysiological mechanism of the impaired cognitive functions. Specifically, we find that the asymmetry of brain operation for epileptic patients is markedly enhanced in temporal lobe and limbic system, in comparison with healthy individuals. The present study indicates that with specifically designed informative neuroimaging markers, resting-state fMRI can serve as a most promising tool for clinical diagnosis, and also shed light onto the physiology behind complex neuropsychiatric disorders. The systematic approaches we present here are expected to have wider applications in general neuropsychiatric disorders.

Annual research review: progress in using brain morphometry as a clinical tool for diagnosing psychiatric disorders

Brain morphometry in recent decades has increased our understanding of the neural bases of psychiatric disorders by localizing anatomical disturbances to specific nuclei and subnuclei of the brain. At least some of these disturbances precede the overt expression of clinical symptoms and possibly are endophenotypes that could be used to diagnose an individual accurately as having a specific psychiatric disorder. More accurate diagnoses could significantly reduce the emotional and financial burden of disease by aiding clinicians in implementing appropriate treatments earlier and in tailoring treatment to the individual needs. Several methods, especially those based on machine learning, have been proposed that use anatomical brain measures and gold-standard diagnoses of participants to learn decision rules that classify a person automatically as having one disorder rather than another. We review the general principles and procedures for machine learning, particularly as applied to diagnostic classification, and then review the procedures that have thus far attempted to diagnose psychiatric illnesses automatically using anatomical measures of the brain. We discuss the strengths and limitations of extant procedures and note that the sensitivity and specificity of these procedures in their most successful implementations have approximated 90%. Although these methods have not yet been applied within clinical settings, they provide strong evidence that individual patients can be diagnosed accurately using the spatial pattern of disturbances across the brain.

Ensemble sparse classification of Alzheimer's disease

The high-dimensional pattern classification methods, e.g., support vector machines (SVM), have been widely investigated for analysis of structural and functional brain images (such as magnetic resonance imaging (MRI)) to assist the diagnosis of Alzheimer's disease (AD) including its prodromal stage, i.e., mild cognitive impairment (MCI). Most existing classification methods extract features from neuroimaging data and then construct a single classifier to perform classification. However, due to noise and small sample size of neuroimaging data, it is challenging to train only a global classifier that can be robust enough to achieve good classification performance. In this paper, instead of building a single global classifier, we propose a local patch-based subspace ensemble method which builds multiple individual classifiers based on different subsets of local patches and then combines them for more accurate and robust classification. Specifically, to capture the local spatial consistency, each brain image is partitioned into a number of local patches and a subset of patches is randomly selected from the patch pool to build a weak classifier. Here, the sparse representation-based classifier (SRC) method, which has shown to be effective for classification of image data (e.g., face), is used to construct each weak classifier. Then, multiple weak classifiers are combined to make the final decision. We evaluate our method on 652 subjects (including 198 AD patients, 225 MCI and 229 normal controls) from Alzheimer's Disease Neuroimaging Initiative (ADNI) database using MR images. The experimental results show that our method achieves an accuracy of 90.8% and an area under the ROC curve (AUC) of 94.86% for AD classification and an accuracy of 87.85% and an AUC of 92.90% for MCI classification, respectively, demonstrating a very promising performance of our method compared with the state-of-the-art methods for AD/MCI classification using MR images.

Using support vector machines with multiple indices of diffusion for automated classification of mild cognitive impairment

Few studies have looked at the potential of using diffusion tensor imaging (DTI) in conjunction with machine learning algorithms in order to automate the classification of healthy older subjects and subjects with mild cognitive impairment (MCI). Here we apply DTI to 40 healthy older subjects and 33 MCI subjects in order to derive values for multiple indices of diffusion within the white matter voxels of each subject. DTI measures were then used together with support vector machines (SVMs) to classify control and MCI subjects. Greater than 90% sensitivity and specificity was achieved using this method, demonstrating the potential of a joint DTI and SVM pipeline for fast, objective classification of healthy older and MCI subjects. Such tools may be useful for large scale drug trials in Alzheimer's disease where the early identification of subjects with MCI is critical.

Individual subject classification for Alzheimer's disease based on incremental learning using a spatial frequency representation of cortical thickness data

Patterns of brain atrophy measured by magnetic resonance structural imaging have been utilized as significant biomarkers for diagnosis of Alzheimer's disease (AD). However, brain atrophy is variable across patients and is non-specific for AD in general. Thus, automatic methods for AD classification require a large number of structural data due to complex and variable patterns of brain atrophy. In this paper, we propose an incremental method for AD classification using cortical thickness data. We represent the cortical thickness data of a subject in terms of their spatial frequency components, employing the manifold harmonic transform. The basis functions for this transform are obtained from the eigenfunctions of the Laplace-Beltrami operator, which are dependent only on the geometry of a cortical surface but not on the cortical thickness defined on it. This facilitates individual subject classification based on incremental learning. In general, methods based on region-wise features poorly reflect the detailed spatial variation of cortical thickness, and those based on vertex-wise features are sensitive to noise. Adopting a vertex-wise cortical thickness representation, our method can still achieve robustness to noise by filtering out high frequency components of the cortical thickness data while reflecting their spatial variation. This compromise leads to high accuracy in AD classification. We utilized MR volumes provided by Alzheimer's Disease Neuroimaging Initiative (ADNI) to validate the performance of the method. Our method discriminated AD patients from Healthy Control (HC) subjects with 82% sensitivity and 93% specificity. It also discriminated Mild Cognitive Impairment (MCI) patients, who converted to AD within 18 months, from non-converted MCI subjects with 63% sensitivity and 76% specificity. Moreover, it showed that the entorhinal cortex was the most discriminative region for classification, which is consistent with previous pathological findings. In comparison with other classification methods, our method demonstrated high classification performance in both categories, which supports the discriminative power of our method in both AD diagnosis and AD prediction.

Using Support Vector Machine to identify imaging biomarkers of neurological and psychiatric disease: a critical review

Standard univariate analysis of neuroimaging data has revealed a host of neuroanatomical and functional differences between healthy individuals and patients suffering a wide range of neurological and psychiatric disorders. Significant only at group level however these findings have had limited clinical translation, and recent attention has turned toward alternative forms of analysis, including Support-Vector-Machine (SVM). A type of machine learning, SVM allows categorisation of an individual's previously unseen data into a predefined group using a classification algorithm, developed on a training data set. In recent years, SVM has been successfully applied in the context of disease diagnosis, transition prediction and treatment prognosis, using both structural and functional neuroimaging data. Here we provide a brief overview of the method and review those studies that applied it to the investigation of Alzheimer's disease, schizophrenia, major depression, bipolar disorder, presymptomatic Huntington's disease, Parkinson's disease and autistic spectrum disorder. We conclude by discussing the main theoretical and practical challenges associated with the implementation of this method into the clinic and possible future directions.