Combining anatomical, diffusion, and resting state functional magnetic resonance imaging for individual classification of mild and moderate Alzheimer's disease

Multimodal classification of drug-naïve first-episode schizophrenia combining anatomical, diffusion and resting state functional resonance imaging

The accurate diagnosis in the early stage of schizophrenia (SZ) is of great importance yet remains challenging. The classification between SZ and control groups based on magnetic resonance imaging (MRI) data using machine learning method could be helpful for SZ diagnosis. Increasing evidence showed that the combination of multimodal MRI data might further improve the classification performance However, medication effect has a profound influence on patients' anatomical and functional features and may reduce the classification efficiency. In this paper, we proposed a multimodal classification method to discriminate drug-naïve first-episode schizophrenia patients from healthy controls (HCs) by a combined structural MRI, diffusion tensor imaging (DTI) and resting state-functional MRI data. To reduce the feature dimension of multimodal data, we applied sparse coding (SC) for feature selection and multi-kernel support vector machine (SVM) for feature combination and classification. The best classification performance with the classification accuracy of 84.29% and area under the receiver operating characteristic (ROC) curve (AUC) of 81.64% was achieved when all modality data were combined. Interestingly, the identified functional markers were mainly found in default mode network (DMN) and cerebellar connections, while the structural markers were within limbic system and prefrontal-thalamo-hippocampal circuit.

The relevance of hippocampal subfield integrity and clock drawing test performance for the diagnosis of Alzheimer's disease and mild cognitive impairment

OBJECTIVES

The clock drawing test (CDT) is one of the worldwide most used screening tests for Alzheimer's disease (AD). MRI studies have identified temporo-parietal regions being involved in CDT impairment. However, the contributions of specific hippocampal subfields and adjacent extrahippocampal structures to CDT performance in AD and mild cognitive impairment (MCI) have not been investigated so far. It is unclear whether morphological alterations or CDT score, or a combination of both, are able to predict AD.

METHODS

38 AD patients, 38 MCI individuals and 31 healthy controls underwent neuropsychological assessment and MRI at 3 Tesla. FreeSurfer 5.3 was used to perform hippocampal parcellation. We used a collection of statistical methods to better understand the relationship between CDT and hippocampal formation. We also tested the clinical feasibility of this relationship when predicting AD.

RESULTS

Impaired CDT performance in AD was associated with widespread atrophy of the cornu ammonis, presubiculum, and subiculum, whereas MCI subjects showed CDT-related alterations of the CA4-dentate gyrus and subiculum. CDT correlates in AD and MCI showed regional and quantitative overlap. Importantly, CDT score was the best predictor of AD.

CONCLUSIONS

Our findings lend support for an involvement of different hippocampal subfields in impaired CDT performance in AD and MCI. CDT seems to be more efficient than subfield imaging for predicting AD.

Single-subject classification of presymptomatic frontotemporal dementia mutation carriers using multimodal MRI

BACKGROUND

Classification models based on magnetic resonance imaging (MRI) may aid early diagnosis of frontotemporal dementia (FTD) but have only been applied in established FTD cases. Detection of FTD patients in earlier disease stages, such as presymptomatic mutation carriers, may further advance early diagnosis and treatment. In this study, we aim to distinguish presymptomatic FTD mutation carriers from controls on an individual level using multimodal MRI-based classification.

METHODS

Anatomical MRI, diffusion tensor imaging (DTI) and resting-state functional MRI data were collected in 55 presymptomatic FTD mutation carriers (8 microtubule-associated protein Tau, 35 progranulin, and 12 chromosome 9 open reading frame 72) and 48 familial controls. We calculated grey and white matter density features from anatomical MRI scans, diffusivity features from DTI, and functional connectivity features from resting-state functional MRI. These features were applied in a recently introduced multimodal behavioural variant FTD (bvFTD) classification model, and were subsequently used to train and test unimodal and multimodal carrier-control models. Classification performance was quantified using area under the receiver operator characteristic curves (AUC).

RESULTS

The bvFTD model was not able to separate presymptomatic carriers from controls beyond chance level (AUC = 0.582, p = 0.078). In contrast, one unimodal and several multimodal carrier-control models performed significantly better than chance level. The unimodal model included the radial diffusivity feature and had an AUC of 0.642 (p = 0.032). The best multimodal model combined radial diffusivity and white matter density features (AUC = 0.684, p = 0.004).

CONCLUSIONS

FTD mutation carriers can be separated from controls with a modest AUC even before symptom-onset, using a newly created carrier-control classification model, while this was not possible using a recent bvFTD classification model. A multimodal MRI-based classification score may therefore be a useful biomarker to aid earlier FTD diagnosis. The exclusive selection of white matter features in the best performing model suggests that the earliest FTD-related pathological processes occur in white matter.

A study of within-subject reliability of the brain's default-mode network

Objective

Resting-state functional magnetic resonance imaging (fMRI) is promising for Alzheimer’s disease (AD). This study aimed to examine short-term reliability of the default-mode network (DMN), one of the main haemodynamic patterns of the brain.

Materials and methods

Using a 1.5 T Philips Achieva scanner, two consecutive resting-state fMRI runs were acquired on 69 healthy adults, 62 patients with mild cognitive impairment (MCI) due to AD, and 28 patients with AD dementia. The anterior and posterior DMN and, as control, the visual-processing network (VPN) were computed using two different methodologies: connectivity of predetermined seeds (theory-driven) and dual regression (data-driven). Divergence and convergence in network strength and topography were calculated with paired t tests, global correlation coefficients, voxel-based correlation maps, and indices of reliability.

Results

No topographical differences were found in any of the networks. High correlations and reliability were found in the posterior DMN of healthy adults and MCI patients. Lower reliability was found in the anterior DMN and in the VPN, and in the posterior DMN of dementia patients.

Discussion

Strength and topography of the posterior DMN appear relatively stable and reliable over a short-term period of acquisition but with some degree of variability across clinical samples.

Sensorineural hearing loss and cognitive impairments: Contributions of thalamus using multiparametric MRI

BACKGROUND

The thalamus is an integrative hub conveying sensory information between cortical areas and related to cognition. However, alterations of the thalamus following partial hearing deprivation remains unknown.

PURPOSE

To investigate the modifications of the thalamus and its seven subdivisions in terms of structure, function, and perfusion in subjects with sensorineural hearing loss (SNHL), as well as their associations with SNHL-induced cognitive impairments.

STUDY TYPE

Cross-sectional study.

SUBJECTS

Thirty-seven bilateral long-term SNHL patients and 38 well-matched controls.

FIELD STRENGTH

3 T/BOLD, T₁ -weighted imaging, arterial spin labeling (ASL).

ASSESSMENT

Quantitative measurements in the thalamus and subdivisions were obtained, including the relative volume, fractional amplitude of low-frequency fluctuation (fALFF) within slow 5 (0.01-0.027 Hz), slow 4 (0.027-0.073 Hz), and combined frequency (0.01-0.073 Hz), as well as the whole-brain functional connectivity. Twenty-five SNHL patients and 20 controls underwent ASL scanning. Then correlation analysis was computed between all significant changes and cognition tests.

STATISTICAL TESTS

Continuous and categorical variables were compared by independent-sample t-test and chi-square test, respectively. Quantitative MRI measurement comparisons were corrected for multiple comparison, and functional connectivity (FC) analysis used two-sample t-test with false-discovery rate correction. Area under the curve (AUC) in receiver operating characteristic curve analysis was applied to evaluate the power of alterations in differentiating SNHL and controls.

RESULTS

No significant difference in the relative volume and perfusion of seven thalamus subdivisions were observed, but a decrease in fALFF in SNHL. SNHL showed reduced thalamic connectivity with the cerebellum lobule VIII, ventral anterior cingulate cortex, insula, superior temporal gyrus, media temporal gyrus, medial frontal gyrus, Heschl's gyrus, and temporal pole. And some FC abnormalities exhibited positive correlations with cognitive tests and high discriminative power (0.8 < AUC < 1) in two groups.

DATA CONCLUSION

SNHL led to decreased thalamic activity and widespread weakened connectivity with other brain areas.

LEVEL OF EVIDENCE

2 Technical Efficacy Stage: 3 J. Magn. Reson. Imaging 2019;50:787-797.

Radiomic Features of Hippocampal Subregions in Alzheimer's Disease and Amnestic Mild Cognitive Impairment

Alzheimer's disease (AD) is characterized by progressive dementia, especially in episodic memory, and amnestic mild cognitive impairment (aMCI) is associated with a high risk of developing AD. Hippocampal atrophy/shape changes are believed to be the most robust magnetic resonance imaging (MRI) markers for AD and aMCI. Radiomics, a method of texture analysis, can quantitatively examine a large set of features and has previously been successfully applied to evaluate imaging biomarkers for AD. To test whether radiomic features in the hippocampus can be employed for early classification of AD and aMCI, 1692 features from the caudal and head parts of the bilateral hippocampus were extracted from 38 AD patients, 33 aMCI patients and 45 normal controls (NCs). One way analysis of variance (ANOVA) showed that 111 features exhibited statistically significant group differences (P < 0.01, Bonferroni corrected). Among these features, 98 were significantly correlated with Mini-Mental State Examination (MMSE) scores in AD and aMCI subjects (P < 0.01). The support vector machine (SVM) model demonstrated that radiomic features allowed us to distinguish AD from NC with an accuracy of 86.75% (specificity = 88.89% and sensitivity = 84.21%) and an area under curve (AUC) of 0.93. In conclusion, these findings provide evidence showing that radiomic features are beneficial in detecting early cognitive decline, and SVM classification analysis provides encouraging evidence for using hippocampal radiomic features as a potential biomarker for clinical applications in AD.

Classification and Prediction of Brain Disorders Using Functional Connectivity: Promising but Challenging

Brain functional imaging data, especially functional magnetic resonance imaging (fMRI) data, have been employed to reflect functional integration of the brain. Alteration in brain functional connectivity (FC) is expected to provide potential biomarkers for classifying or predicting brain disorders. In this paper, we present a comprehensive review in order to provide guidance about the available brain FC measures and typical classification strategies. We survey the state-of-the-art FC analysis methods including widely used static functional connectivity (SFC) and more recently proposed dynamic functional connectivity (DFC). Temporal correlations among regions of interest (ROIs), data-driven spatial network and functional network connectivity (FNC) are often computed to reflect SFC from different angles. SFC can be extended to DFC using a sliding-window framework, and intrinsic connectivity states along the time-varying connectivity patterns are typically extracted using clustering or decomposition approaches. We also briefly summarize window-less DFC approaches. Subsequently, we highlight various strategies for feature selection including the filter, wrapper and embedded methods. In terms of model building, we include traditional classifiers as well as more recently applied deep learning methods. Moreover, we review representative applications with remarkable classification accuracy for psychosis and mood disorders, neurodevelopmental disorder, and neurological disorders using fMRI data. Schizophrenia, bipolar disorder, autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), Alzheimer's disease and mild cognitive impairment (MCI) are discussed. Finally, challenges in the field are pointed out with respect to the inaccurate diagnosis labeling, the abundant number of possible features and the difficulty in validation. Some suggestions for future work are also provided.

Single-subject classification of presymptomatic frontotemporal dementia mutation carriers using multimodal MRI

Background

Classification models based on magnetic resonance imaging (MRI) may aid early diagnosis of frontotemporal dementia (FTD) but have only been applied in established FTD cases. Detection of FTD patients in earlier disease stages, such as presymptomatic mutation carriers, may further advance early diagnosis and treatment. In this study, we aim to distinguish presymptomatic FTD mutation carriers from controls on an individual level using multimodal MRI-based classification.

Methods

Anatomical MRI, diffusion tensor imaging (DTI) and resting-state functional MRI data were collected in 55 presymptomatic FTD mutation carriers (8 microtubule-associated protein Tau, 35 progranulin, and 12 chromosome 9 open reading frame 72) and 48 familial controls. We calculated grey and white matter density features from anatomical MRI scans, diffusivity features from DTI, and functional connectivity features from resting-state functional MRI. These features were applied in a recently introduced multimodal behavioural variant FTD (bvFTD) classification model, and were subsequently used to train and test unimodal and multimodal carrier-control models. Classification performance was quantified using area under the receiver operator characteristic curves (AUC).

Results

The bvFTD model was not able to separate presymptomatic carriers from controls beyond chance level (AUC = 0.570, p = 0.11). In contrast, one unimodal and several multimodal carrier-control models performed significantly better than chance level. The unimodal model included the radial diffusivity feature and had an AUC of 0.646 (p = 0.021). The best multimodal model combined radial diffusivity and white matter density features (AUC = 0.680, p = 0.005).

Conclusions

FTD mutation carriers can be separated from controls with a modest AUC even before symptom-onset, using a newly created carrier-control classification model, while this was not possible using a recent bvFTD classification model. A multimodal MRI-based classification score may therefore be a useful biomarker to aid earlier FTD diagnosis. The exclusive selection of white matter features in the best performing model suggests that the earliest FTD-related pathological processes occur in white matter.

Quasi-Periodic Patterns of Neural Activity improve Classification of Alzheimer's Disease in Mice

Resting state (rs)fMRI allows measurement of brain functional connectivity and has identified default mode (DMN) and task positive (TPN) network disruptions as promising biomarkers for Alzheimer's disease (AD). Quasi-periodic patterns (QPPs) of neural activity describe recurring spatiotemporal patterns that display DMN with TPN anti-correlation. We reasoned that QPPs could provide new insights into AD network dysfunction and improve disease diagnosis. We therefore used rsfMRI to investigate QPPs in old TG2576 mice, a model of amyloidosis, and age-matched controls. Multiple QPPs were determined and compared across groups. Using linear regression, we removed their contribution from the functional scans and assessed how they reflected functional connectivity. Lastly, we used elastic net regression to determine if QPPs improved disease classification. We present three prominent findings: (1) Compared to controls, TG2576 mice were marked by opposing neural dynamics in which DMN areas were anti-correlated and displayed diminished anti-correlation with the TPN. (2) QPPs reflected lowered DMN functional connectivity in TG2576 mice and revealed significantly decreased DMN-TPN anti-correlations. (3) QPP-derived measures significantly improved classification compared to conventional functional connectivity measures. Altogether, our findings provide insight into the neural dynamics of aberrant network connectivity in AD and indicate that QPPs might serve as a translational diagnostic tool.

Detecting Alzheimer's Disease on Small Dataset: A Knowledge Transfer Perspective

Computer-aided diagnosis (CAD) is an attractive topic in Alzheimer's disease (AD) research. Many algorithms are based on a relatively large training dataset. However, small hospitals are usually unable to collect sufficient training samples for robust classification. Although data sharing is expanding in scientific research, it is unclear whether a model based on one dataset is well suited for other data sources. Using a small dataset from a local hospital and a large shared dataset from the AD neuroimaging initiative, we conducted a heterogeneity analysis and found that different functional magnetic resonance imaging data sources show different sample distributions in feature space. In addition, we proposed an effective knowledge transfer method to diminish the disparity among different datasets and improve the classification accuracy on datasets with insufficient training samples. The accuracy increased by approximately 20% compared with that of a model based only on the original small dataset. The results demonstrated that the proposed approach is a novel and effective method for CAD in hospitals with only small training datasets. It solved the challenge of limited sample size in detection of AD, which is a common issue but lack of adequate attention. Furthermore, this paper sheds new light on effective use of multi-source data for neurological disease diagnosis.

Resting-state connectivity in neurodegenerative disorders: Is there potential for an imaging biomarker?

Biomarkers in whichever modality are tremendously important in diagnosing of disease, tracking disease progression and clinical trials. This applies in particular for disorders with a long disease course including pre-symptomatic stages, in which only subtle signs of clinical progression can be observed. Magnetic resonance imaging (MRI) biomarkers hold particular promise due to their relative ease of use, cost-effectiveness and non-invasivity. Studies measuring resting-state functional MR connectivity have become increasingly common during recent years and are well established in neuroscience and related fields. Its increasing application does of course also include clinical settings and therein neurodegenerative diseases. In the present review, we critically summarise the state of the literature on resting-state functional connectivity as measured with functional MRI in neurodegenerative disorders. In addition to an overview of the results, we briefly outline the methods applied to the concept of resting-state functional connectivity. While there are many different neurodegenerative disorders cumulatively affecting a substantial number of patients, for most of them studies on resting-state fMRI are lacking. Plentiful amounts of papers are available for Alzheimer's disease (AD) and Parkinson's disease (PD), but only few works being available for the less common neurodegenerative diseases. This allows some conclusions on the potential of resting-state fMRI acting as a biomarker for the aforementioned two diseases, but only tentative statements for the others. For AD, the literature contains a relatively strong consensus regarding an impairment of the connectivity of the default mode network compared to healthy individuals. However, for AD there is no considerable documentation on how that alteration develops longitudinally with the progression of the disease. For PD, the available research points towards alterations of connectivity mainly in limbic and motor related regions and networks, but drawing conclusions for PD has to be done with caution due to a relative heterogeneity of the disease. For rare neurodegenerative diseases, no clear conclusions can be drawn due to the few published results. Nevertheless, summarising available data points towards characteristic connectivity alterations in Huntington's disease, frontotemporal dementia, dementia with Lewy bodies, multiple systems atrophy and the spinocerebellar ataxias. Overall at this point in time, the data on AD are most promising towards the eventual use of resting-state fMRI as an imaging biomarker, although there remain issues such as reproducibility of results and a lack of data demonstrating longitudinal changes. Improved methods providing more precise classifications as well as resting-state network changes that are sensitive to disease progression or therapeutic intervention are highly desirable, before routine clinical use could eventually become a reality.

The Added Value of Diffusion-Weighted MRI-Derived Structural Connectome in Evaluating Mild Cognitive Impairment: A Multi-Cohort Validation1

T1-weighted MRI has been extensively used to extract imaging biomarkers and build classification models for differentiating Alzheimer's disease (AD) patients from healthy controls, but only recently have brain connectome networks derived from diffusion-weighted MRI been used to model AD progression and various stages of disease such as mild cognitive impairment (MCI). MCI, as a possible prodromal stage of AD, has gained intense interest recently, since it may be used to assess risk factors for AD. Little work has been done to combine information from both white matter and gray matter, and it is unknown how much classification power the diffusion-weighted MRI-derived structural connectome could provide beyond information available from T1-weighted MRI. In this paper, we focused on investigating whether diffusion-weighted MRI-derived structural connectome can improve differentiating healthy controls subjects from those with MCI. Specifically, we proposed a novel feature-ranking method to build classification models using the most highly ranked feature variables to classify MCI with healthy controls. We verified our method on two independent cohorts including the second stage of Alzheimer's Disease Neuroimaging Initiative (ADNI2) database and the National Alzheimer's Coordinating Center (NACC) database. Our results indicated that 1) diffusion-weighted MRI-derived structural connectome can complement T1-weighted MRI in the classification task; 2) the feature-rank method is effective because of the identified consistent T1-weighted MRI and network feature variables on ADNI2 and NACC. Furthermore, by comparing the top-ranked feature variables from ADNI2, NACC, and combined dataset, we concluded that cross-validation using independent cohorts is necessary and highly recommended.

Factors influencing accuracy of cortical thickness in the diagnosis of Alzheimer's disease

There is great value to use of structural neuroimaging in the assessment of Alzheimer's disease (AD). However, to date, predictive value of structural imaging tend to range between 80% and 90% in accuracy and it is unclear why this is the case given that structural imaging should parallel the pathologic processes of AD. There is a possibility that clinical misdiagnosis relative to the gold standard pathologic diagnosis and/or additional brain pathologies are confounding factors contributing to reduced structural imaging classification accuracy. We examined potential factors contributing to misclassification of individuals with clinically diagnosed AD purely from cortical thickness measures. Correctly classified and incorrectly classified groups were compared across a range of demographic, biological, and neuropsychological data including cerebrospinal fluid biomarkers, amyloid imaging, white matter hyperintensity (WMH) volume, cognitive, and genetic factors. Individual subject analyses suggested that at least a portion of the control individuals misclassified as AD from structural imaging additionally harbor substantial AD biomarker pathology and risk, yet are relatively resistant to cognitive symptoms, likely due to "cognitive reserve," and therefore clinically unimpaired. In contrast, certain clinical control individuals misclassified as AD from cortical thickness had increased WMH volume relative to other controls in the sample, suggesting that vascular conditions may contribute to classification accuracy from cortical thickness measures. These results provide examples of factors that contribute to the accuracy of structural imaging in predicting a clinical diagnosis of AD, and provide important information about considerations for future work aimed at optimizing structural based diagnostic classifiers for AD.

Random forest feature selection, fusion and ensemble strategy: Combining multiple morphological MRI measures to discriminate among healhy elderly, MCI, cMCI and alzheimer's disease patients: From the alzheimer's disease neuroimaging initiative (ADNI) database

BACKGROUND

In the era of computer-assisted diagnostic tools for various brain diseases, Alzheimer's disease (AD) covers a large percentage of neuroimaging research, with the main scope being its use in daily practice. However, there has been no study attempting to simultaneously discriminate among Healthy Controls (HC), early mild cognitive impairment (MCI), late MCI (cMCI) and stable AD, using features derived from a single modality, namely MRI.

NEW METHOD

Based on preprocessed MRI images from the organizers of a neuroimaging challenge,³ we attempted to quantify the prediction accuracy of multiple morphological MRI features to simultaneously discriminate among HC, MCI, cMCI and AD. We explored the efficacy of a novel scheme that includes multiple feature selections via Random Forest from subsets of the whole set of features (e.g. whole set, left/right hemisphere etc.), Random Forest classification using a fusion approach and ensemble classification via majority voting. From the ADNI database, 60 HC, 60 MCI, 60 cMCI and 60 CE were used as a training set with known labels. An extra dataset of 160 subjects (HC: 40, MCI: 40, cMCI: 40 and AD: 40) was used as an external blind validation dataset to evaluate the proposed machine learning scheme.

RESULTS

In the second blind dataset, we succeeded in a four-class classification of 61.9% by combining MRI-based features with a Random Forest-based Ensemble Strategy. We achieved the best classification accuracy of all teams that participated in this neuroimaging competition.

COMPARISON WITH EXISTING METHOD(S)

The results demonstrate the effectiveness of the proposed scheme to simultaneously discriminate among four groups using morphological MRI features for the very first time in the literature.

CONCLUSIONS

Hence, the proposed machine learning scheme can be used to define single and multi-modal biomarkers for AD.

Looking for Alzheimer's Disease morphometric signatures using machine learning techniques

BACKGROUND

We present our results in the International challenge for automated prediction of MCI from MRI data. We evaluate the performance of MRI-based neuromorphometrics features (nMF) in the classification of Healthy Controls (HC), Mild Cognitive Impairment (MCI), converters MCI (cMCI) and Alzheimer's Disease (AD) patients.

NEW METHODS

We propose to segregate participants in three groups according to Mini Mental State Examination score (MMSEs), searching for the main nMF in each group. Then we use them to develop a Multi Classifier System (MCS). We compare the MCS against a single classifier scheme using both MMSEs+nMF and nMF only. We repeat this comparison using three state-of-the-art classification algorithms.

RESULTS

The MCS showed the best performance on both Accuracy and Area Under the Receiver Operating Curve (AUC) in comparison with single classifiers. The multiclass AUC for the MCS classification on Test Dataset were 0.83 for HC, 0.76 for cMCI, 0.65 for MCI and 0.95 for AD. Furthermore, MCS's optimum accuracy on Neurodegenerative Disease (ND) detection (AD+cMCI vs MCI+HC) was 81.0% (AUC=0.88), while the single classifiers got 71.3% (AUC=0.86) and 63.1% (AUC=0.79) for MMSEs+nMF and only nMF respectively.

COMPARISON WITH EXISTING METHOD

The proposed MCS showed a better performance than using all nMF into a single state-of-the-art classifier.

CONCLUSIONS

These findings suggest that using cognitive scoring, e.g. MMSEs, in the design of a Multi Classifier System improves performance by allowing a better selection of MRI-based features.

A comprehensive analysis of resting state fMRI measures to classify individual patients with Alzheimer's disease

Alzheimer's disease (AD) patients show altered patterns of functional connectivity (FC) on resting state functional magnetic resonance imaging (RSfMRI) scans. It is yet unclear which RSfMRI measures are most informative for the individual classification of AD patients. We investigated this using RSfMRI scans from 77 AD patients (MMSE = 20.4 ± 4.5) and 173 controls (MMSE = 27.5 ± 1.8). We calculated i) FC matrices between resting state components as obtained with independent component analysis (ICA), ii) the dynamics of these FC matrices using a sliding window approach, iii) the graph properties (e.g., connection degree, and clustering coefficient) of the FC matrices, and iv) we distinguished five FC states and administered how long each subject resided in each of these five states. Furthermore, for each voxel we calculated v) FC with 10 resting state networks using dual regression, vi) FC with the hippocampus, vii) eigenvector centrality, and viii) the amplitude of low frequency fluctuations (ALFF). These eight measures were used separately as predictors in an elastic net logistic regression, and combined in a group lasso logistic regression model. We calculated the area under the receiver operating characteristic curve plots (AUC) to determine classification performance. The AUC values ranged between 0.51 and 0.84 and the highest were found for the FC matrices (0.82), FC dynamics (0.84) and ALFF (0.82). The combination of all measures resulted in an AUC of 0.85. We show that it is possible to obtain moderate to good AD classification using RSfMRI scans. FC matrices, FC dynamics and ALFF are most discriminative and the combination of all the resting state measures improves classification accuracy slightly.

Ensemble based on static classifier selection for automated diagnosis of Mild Cognitive Impairment

BACKGROUND

Alzheimer's disease (AD) is the most common cause of neurodegenerative dementia in the elderly population. Scientific research is very active in the challenge of designing automated approaches to achieve an early and certain diagnosis. Recently an international competition among AD predictors has been organized: "A Machine learning neuroimaging challenge for automated diagnosis of Mild Cognitive Impairment" (MLNeCh). This competition is based on pre-processed sets of T1-weighted Magnetic Resonance Images (MRI) to be classified in four categories: stable AD, individuals with MCI who converted to AD, individuals with MCI who did not convert to AD and healthy controls.

NEW METHOD

In this work, we propose a method to perform early diagnosis of AD, which is evaluated on MLNeCh dataset. Since the automatic classification of AD is based on the use of feature vectors of high dimensionality, different techniques of feature selection/reduction are compared in order to avoid the curse-of-dimensionality problem, then the classification method is obtained as the combination of Support Vector Machines trained using different clusters of data extracted from the whole training set.

RESULTS

The multi-classifier approach proposed in this work outperforms all the stand-alone method tested in our experiments. The final ensemble is based on a set of classifiers, each trained on a different cluster of the training data. The proposed ensemble has the great advantage of performing well using a very reduced version of the data (the reduction factor is more than 90%). The MATLAB code for the ensemble of classifiers will be publicly available¹ to other researchers for future comparisons.

Distinct resting-state functional connections associated with episodic and visuospatial memory in older adults

Episodic and spatial memory are commonly impaired in ageing and Alzheimer's disease. Volumetric and task-based functional magnetic resonance imaging (fMRI) studies suggest a preferential involvement of the medial temporal lobe (MTL), particularly the hippocampus, in episodic and spatial memory processing. The present study examined how these two memory types were related in terms of their associated resting-state functional architecture. 3T multiband resting state fMRI scans from 497 participants (60–82 years old) of the cross-sectional Whitehall II Imaging sub-study were analysed using an unbiased, data-driven network-modelling technique (FSLNets). Factor analysis was performed on the cognitive battery; the Hopkins Verbal Learning test and Rey-Osterreith Complex Figure test factors were used to assess verbal and visuospatial memory respectively. We present a map of the macroscopic functional connectome for the Whitehall II Imaging sub-study, comprising 58 functionally distinct nodes clustered into five major resting-state networks. Within this map we identified distinct functional connections associated with verbal and visuospatial memory. Functional anticorrelation between the hippocampal formation and the frontal pole was significantly associated with better verbal memory in an age-dependent manner. In contrast, hippocampus–motor and parietal–motor functional connections were associated with visuospatial memory independently of age. These relationships were not driven by grey matter volume and were unique to the respective memory domain. Our findings provide new insights into current models of brain-behaviour interactions, and suggest that while both episodic and visuospatial memory engage MTL nodes of the default mode network, the two memory domains differ in terms of the associated functional connections between the MTL and other resting-state brain networks.

•

Episodic and visuospatial memory engaged a common medial temporal lobe substrate at rest.

•

However, the resting-state functional connections of the MTL differed based on the memory demand.

•

Visuospatial memory was associated with hippocampal-parietal and motorparietal interaction.

•

Verbal memory was associated with hippocampus-frontal pole anticorrelation.

•

Findings provide novel insights into resting-state brain-behaviour interactions in older adults.

A review on neuroimaging-based classification studies and associated feature extraction methods for Alzheimer's disease and its prodromal stages

Neuroimaging has made it possible to measure pathological brain changes associated with Alzheimer's disease (AD) in vivo. Over the past decade, these measures have been increasingly integrated into imaging signatures of AD by means of classification frameworks, offering promising tools for individualized diagnosis and prognosis. We reviewed neuroimaging-based studies for AD and mild cognitive impairment classification, selected after online database searches in Google Scholar and PubMed (January, 1985-June, 2016). We categorized these studies based on the following neuroimaging modalities (and sub-categorized based on features extracted as a post-processing step from these modalities): i) structural magnetic resonance imaging [MRI] (tissue density, cortical surface, and hippocampal measurements), ii) functional MRI (functional coherence of different brain regions, and the strength of the functional connectivity), iii) diffusion tensor imaging (patterns along the white matter fibers), iv) fluorodeoxyglucose positron emission tomography (FDG-PET) (metabolic rate of cerebral glucose), and v) amyloid-PET (amyloid burden). The studies reviewed indicate that the classification frameworks formulated on the basis of these features show promise for individualized diagnosis and prediction of clinical progression. Finally, we provided a detailed account of AD classification challenges and addressed some future research directions.

Quantifying the Interaction and Contribution of Multiple Datasets in Fusion: Application to the Detection of Schizophrenia

The extraction of information from multiple sets of data is a problem inherent to many disciplines. This is possible by either analyzing the data sets jointly as in data fusion or separately and then combining as in data integration. However, selecting the optimal method to combine and analyze multiset data is an ever-present challenge. The primary reason for this is the difficulty in determining the optimal contribution of each data set to an analysis as well as the amount of potentially exploitable complementary information among data sets. In this paper, we propose a novel classification rate-based technique to unambiguously quantify the contribution of each data set to a fusion result as well as facilitate direct comparisons of fusion methods on real data and apply a new method, independent vector analysis (IVA), to multiset fusion. This classification rate-based technique is used on functional magnetic resonance imaging data collected from 121 patients with schizophrenia and 150 healthy controls during the performance of three tasks. Through this application, we find that though optimal performance is achieved by exploiting all tasks, each task does not contribute equally to the result and this framework enables effective quantification of the value added by each task. Our results also demonstrate that data fusion methods are more powerful than data integration methods, with the former achieving a classification rate of 73.5 % and the latter achieving one of 70.9 %, a difference which we show is significant when all three tasks are analyzed together. Finally, we show that IVA, due to its flexibility, has equivalent or superior performance compared with the popular data fusion method, joint independent component analysis.

Dissociable diffusion MRI patterns of white matter microstructure and connectivity in Alzheimer's disease spectrum

Recent efforts using diffusion tensor imaging (DTI) have documented white matter (WM) alterations in Alzheimer's disease (AD). The full potential of whole-brain DTI, however, has not been fully exploited as studies have focused on individual microstructural indices independently. In patients with AD (n = 79), mild (MCI, n = 55) and subjective (SCI, n = 30) cognitive impairment, we applied linked independent component analysis (LICA) to model inter-subject variability across five complementary DTI measures (fractional anisotropy (FA), axial/radial/mean diffusivity, diffusion tensor mode), two crossing fiber measures estimated using a multi-compartment crossing-fiber model reflecting the volume fraction of the dominant (f1) and non-dominant (f2) diffusion orientation, and finally, connectivity density obtained from full-brain probabilistic tractography. The LICA component explaining the largest data variance was highly sensitive to disease severity (AD < MCI < SCI) and revealed widespread coordinated decreases in FA and f1 with increases in all diffusivity measures in AD. Additionally, it reflected regional coordinated decreases and increases in f2, mode and connectivity density, implicating bidirectional alterations of crossing fibers in the fornix, uncinate fasciculi, corpus callosum and major sensorimotor pathways. LICA yielded improved diagnostic classification performance compared to univariate region-of-interest features. Our results document coordinated WM microstructural and connectivity alterations in line with disease severity across the AD continuum.

Individual classification of Alzheimer's disease with diffusion magnetic resonance imaging

Diffusion magnetic resonance imaging (MRI) is a powerful non-invasive method to study white matter integrity, and is sensitive to detect differences in Alzheimer's disease (AD) patients. Diffusion MRI may be able to contribute towards reliable diagnosis of AD. We used diffusion MRI to classify AD patients (N=77), and controls (N=173). We use different methods to extract information from the diffusion MRI data. First, we use the voxel-wise diffusion tensor measures that have been skeletonised using tract based spatial statistics. Second, we clustered the voxel-wise diffusion measures with independent component analysis (ICA), and extracted the mixing weights. Third, we determined structural connectivity between Harvard Oxford atlas regions with probabilistic tractography, as well as graph measures based on these structural connectivity graphs. Classification performance for voxel-wise measures ranged between an AUC of 0.888, and 0.902. The ICA-clustered measures ranged between an AUC of 0.893, and 0.920. The AUC for the structural connectivity graph was 0.900, while graph measures based upon this graph ranged between an AUC of 0.531, and 0.840. All measures combined with a sparse group lasso resulted in an AUC of 0.896. Overall, fractional anisotropy clustered into ICA components was the best performing measure. These findings may be useful for future incorporation of diffusion MRI into protocols for AD classification, or as a starting point for early detection of AD using diffusion MRI.

Robust Detection of Impaired Resting State Functional Connectivity Networks in Alzheimer's Disease Using Elastic Net Regularized Regression

The large number of multicollinear regional features that are provided by resting state (rs) fMRI data requires robust feature selection to uncover consistent networks of functional disconnection in Alzheimer's disease (AD). Here, we compared elastic net regularized and classical stepwise logistic regression in respect to consistency of feature selection and diagnostic accuracy using rs-fMRI data from four centers of the “German resting-state initiative for diagnostic biomarkers” (psymri.org), comprising 53 AD patients and 118 age and sex matched healthy controls. Using all possible pairs of correlations between the time series of rs-fMRI signal from 84 functionally defined brain regions as the initial set of predictor variables, we calculated accuracy of group discrimination and consistency of feature selection with bootstrap cross-validation. Mean areas under the receiver operating characteristic curves as measure of diagnostic accuracy were 0.70 in unregularized and 0.80 in regularized regression. Elastic net regression was insensitive to scanner effects and recovered a consistent network of functional connectivity decline in AD that encompassed parts of the dorsal default mode as well as brain regions involved in attention, executive control, and language processing. Stepwise logistic regression found no consistent network of AD related functional connectivity decline. Regularized regression has high potential to increase diagnostic accuracy and consistency of feature selection from multicollinear functional neuroimaging data in AD. Our findings suggest an extended network of functional alterations in AD, but the diagnostic accuracy of rs-fMRI in this multicenter setting did not reach the benchmark defined for a useful biomarker of AD.

Diagnostic classification of unipolar depression based on resting-state functional connectivity MRI: effects of generalization to a diverse sample

In small, selected samples, an approach combining resting-state functional connectivity MRI and multivariate pattern analysis has been able to successfully classify patients diagnosed with unipolar depression. Purposes of this investigation were to assess the generalizability of this approach to a large clinically more realistic sample and secondarily to assess the replicability of previously reported methodological feasibility in a more homogeneous subgroup with pronounced depressive symptoms. Two independent subsets were drawn from the depression and control cohorts of the BiDirect study, each with 180 patients with and 180 controls without depression. Functional connectivity either among regions covering the gray matter or selected regions with known alterations in depression was assessed by resting-state fMRI. Support vector machines with and without automated feature selection were used to train classifiers differentiating between individual patients and controls in the entire first subset as well as in the subgroup. Model parameters were explored systematically. The second independent subset was used for validation of successful models. Classification accuracies in the large, heterogeneous sample ranged from 45.0 to 56.1% (chance level 50.0%). In the subgroup with higher depression severity, three out of 90 models performed significantly above chance (60.8-61.7% at independent validation). In conclusion, common classification methods previously successful in small homogenous depression samples do not immediately translate to a more realistic population. Future research to develop diagnostic classification approaches in depression should focus on more specific clinical questions and consider heterogeneity, including symptom severity as an important factor.

Combining multiple anatomical MRI measures improves Alzheimer's disease classification

Several anatomical MRI markers for Alzheimer's disease (AD) have been identified. Hippocampal volume, cortical thickness, and grey matter density have been used successfully to discriminate AD patients from controls. These anatomical MRI measures have so far mainly been used separately. The full potential of anatomical MRI scans for AD diagnosis might thus not yet have been used optimally. In this study, we therefore combined multiple anatomical MRI measures to improve diagnostic classification of AD. For 21 clinically diagnosed AD patients and 21 cognitively normal controls, we calculated (i) cortical thickness, (ii) cortical area, (iii) cortical curvature, (iv) grey matter density, (v) subcortical volumes, and (vi) hippocampal shape. These six measures were used separately and combined as predictors in an elastic net logistic regression. We made receiver operating curve plots and calculated the area under the curve (AUC) to determine classification performance. AUC values for the single measures ranged from 0.67 (cortical thickness) to 0.94 (grey matter density). The combination of all six measures resulted in an AUC of 0.98. Our results demonstrate that the different anatomical MRI measures contain complementary information. A combination of these measures may therefore improve accuracy of AD diagnosis in clinical practice. Hum Brain Mapp 37:1920-1929, 2016. © 2016 Wiley Periodicals, Inc.