Harini Eavani

GraSP: geodesic Graph-based Segmentation with Shape Priors for the functional parcellation of the cortex.

Resting-state functional MRI is a powerful technique for mapping the functional organization of the human brain. However, for many types of connectivity analysis, high-resolution voxelwise analyses are computationally infeasible and dimensionality reduction is typically used to limit the number of network nodes. Most commonly, network nodes are defined using standard anatomic atlases that do not align well with functional neuroanatomy or regions of interest covering a small portion of the cortex. Data-driven parcellation methods seek to overcome such limitations, but existing approaches are highly dependent on initialization procedures and produce spatially fragmented parcels or overly isotropic parcels that are unlikely to be biologically grounded. In this paper, we propose a novel graph-based parcellation method that relies on a discrete Markov Random Field framework. The spatial connectedness of the parcels is explicitly enforced by shape priors. The shape of the parcels is adapted to underlying data through the use of functional geodesic distances. Our method is initialization-free and rapidly segments the cortex in a single optimization. The performance of the method was assessed using a large developmental cohort of more than 850 subjects. Compared to two prevalent parcellation methods, our approach provides superior reproducibility for a similar data fit. Furthermore, compared to other methods, it avoids incoherent parcels. Finally, the methods utility is demonstrated through its ability to detect strong brain developmental effects that are only weakly observed using other methods.

Sparse Dictionary Learning of Resting State fMRI Networks

Research in resting state fMRI (rsfMRI) has revealed the presence of stable, anti-correlated functional sub-networks in the brain. Task-positive networks are active during a cognitive process and are anti-correlated with task-negative networks, which are active during rest. In this paper, based on the assumption that the structure of the resting state functional brain connectivity is sparse, we utilize sparse dictionary modeling to identify distinct functional sub-networks. We propose two ways of formulating the sparse functional network learning problem that characterize the underlying functional connectivity from different perspectives. Our results show that the whole-brain functional connectivity can be concisely represented with highly modular, overlapping task-positive/negative pairs of sub-networks.

White matter atlas generation using HARDI based automated parcellation.

Most diffusion imaging studies have used subject registration to an atlas space for enhanced quantification of anatomy. However, standard diffusion tensor atlases lack information in regions of fiber crossing and are based on adult anatomy. The degree of error associated with applying these atlases to studies of children for example has not yet been estimated but may lead to suboptimal results. This paper describes a novel technique for generating population-specific high angular resolution diffusion imaging (HARDI)-based atlases consisting of labeled regions of homogenous white matter. Our approach uses a fiber orientation distribution (FOD) diffusion model and a data driven clustering algorithm. White matter regional labeling is achieved by our automated data driven clustering algorithm that has the potential to delineate white matter regions based on fiber complexity and orientation. The advantage of such an atlas is that it is study specific and more comprehensive in describing regions of white matter homogeneity as compared to standard anatomical atlases. We have applied this state of the art technique to a dataset consisting of adolescent and preadolescent children, creating one of the first examples of a HARDI-based atlas, thereby establishing the feasibility of the atlas creation framework. The white matter regions generated by our automated clustering algorithm have lower FOD variance than when compared to the regions created from a standard anatomical atlas.

HARDI based pattern classifiers for the identification of white matter pathologies

The paper presents a method for creating abnormality classifiers from high angular resolution diffusion imaging (HARDI) data. We utilized the fiber orientation distribution (FOD) diffusion model to represent the local WM architecture of each subject. The FOD images are then spatially normalized to a common template using a non-linear registration technique. Regions of homogeneous white matter architecture (ROIs) are determined by applying a parcellation algorithm to the population average FOD image. Orientation invariant features of each ROIs mean FOD are determined and concatenated into a feature vector to represent each subject. Principal component analysis (PCA) was used for dimensionality reduction and a linear support vector machine (SVM) classifier is trained on the PCA coefficients. The classifier assigns each test subject a probabilistic score indicating the likelihood of belonging to the patient group. The method was validated using a 5 fold validation scheme on a population containing autism spectrum disorder (ASD) patients and typically developing (TD) controls. A clear distinction between ASD patients and controls was obtained with a 77% accuracy.

Discriminative sparse connectivity patterns for classification of fMRI Data.

Functional connectivity using resting-state fMRI has emerged as an important research tool for understanding normal brain function as well as changes occurring during brain development and in various brain disorders. Most prior work has examined changes in pairwise functional connectivity values using a multi-variate classification approach, such as Support Vector Machines (SVM). While it is powerful, SVMs produce a dense set of high-dimensional weight vectors as output, which are difficult to interpret, and require additional post-processing to relate to known functional networks. In this paper, we propose a joint framework that combines network identification and classification, resulting in a set of networks, or Sparse Connectivity Patterns (SCPs) which are functionally interpretable as well as highly discriminative of the two groups. Applied to a study of normal development classifying children vs. adults, the proposed method provided accuracy of 76%(AUC= 0.85), comparable to SVM (79%,AUC=0.87), but with dramatically fewer number of features (50 features vs. 34716 for the SVM). More importantly, this leads to a tremendous improvement in neuro-scientific interpretability, which is specially advantageous in such a study where the group differences are wide-spread throughout the brain. Highest-ranked discriminative SCPs reflect increases in long-range connectivity in adults between the frontal areas and posterior cingulate regions. In contrast, connectivity between the bilateral parahippocampal gyri was decreased in adults compared to children.

Identifying patterns in temporal variation of functional connectivity using resting state FMRI

Estimating functional brain networks from fMRI data has been the focus of much research in recent years. Low sample sizes (time-points) and high dimensionality of fMRI has restricted estimation to a temporally averaged connectivity matrix per subject, due to which the dynamics of functional connectivity is largely unknown. In this paper, we propose a novel method based on constrained matrix factorization that addresses two major issues. Firstly, it finds a set of basis networks that are the semantic parts of the time-varying whole-brain functional networks. The whole-brain network at any point in time, for any subject, is a non-negative combination of these basis networks. Secondly, significant dimensionality reduction is achieved by projecting the data onto this basis, facilitating subsequent analysis of temporal dynamics. Results on simulated fMRI data show that our method can effectively recover underlying basis networks. We apply this method on a normative dataset of resting state fMRI scans. Results indicate that the functional connectivity of a subject at any point during the scan is composed of combinations of overlapping task-positive/negative pairs of sub-networks.

A Graph-Based Brain Parcellation Method Extracting Sparse Networks

fMRI is a powerful tool for assessing the functioning of the brain. The analysis of resting-state fMRI allows to describe the functional relationship between the cortical areas. Since most connectivity analysis methods suffer from the curse of dimensionality, the cortex needs to be first partitioned into regions of coherent activation patterns. Once the signals of these regions of interest have been extracted, estimating a sparse approximation of the inverse of their correlation matrix is a classical way to robustly describe their functional interactions. In this paper, we address both objectives with a novel parcellation method based on Markov Random Fields that favors the extraction of sparse networks of regions. Our method relies on state of the art rsfMRI models, naturally adapts the number of parcels to the data and is guaranteed to provide connected regions due to the use of shape priors. The second contribution of this paper resides in two novel sparsity enforcing potentials. Our approach is validated with a publicly available dataset.

Heterogeneity of structural and functional imaging patterns of advanced brain aging revealed via machine learning methods

Disentangling the heterogeneity of brain aging in cognitively normal older adults is challenging, as multiple co-occurring pathologic processes result in diverse functional and structural changes. Capitalizing on machine learning methods applied to magnetic resonance imaging data from 400 participants aged 50 to 96 years in the Baltimore Longitudinal Study of Aging, we constructed normative cross-sectional brain aging trajectories of structural and functional changes. Deviations from typical trajectories identified individuals with resilient brain aging and multiple subtypes of advanced brain aging. We identified 5 distinct phenotypes of advanced brain aging. One group included individuals with relatively extensive structural and functional loss and high white matter hyperintensity burden. Another subgroup showed focal hippocampal atrophy and lower posterior-cingulate functional coherence, low white matter hyperintensity burden, and higher medial-temporal connectivity, potentially reflecting high brain tissue reserve counterbalancing brain loss that is consistent with early stages of Alzheimers disease. Other subgroups displayed distinct patterns. These results indicate that brain changes should not be measured seeking a single signature of brain aging but rather via methods capturing heterogeneity and subtypes of brain aging. Our findings inform future studies aiming to better understand the neurobiological underpinnings of brain aging imaging patterns.

MAPPING THE HETEROGENEITY OF NEUROANATOMY AND FUNCTIONAL CONNECTIVITY DEVIATION FROM TYPICAL BRAIN AGING: A PATTERN ANALYSIS AND MACHINE LEARNING STUDY

(DVR) and compared the relationship between age and global PiB in SA and TOA.Multiple regressionmodels were used to determine relationships between morphometric measures, global PiB, memory and processing speed. Results:Cortical thickness analyses revealed several regions of preserved cortical integrity in SA compared to TOA (p<0.05, uncorrected), including right anterior cingulate and prefrontal cortex, which have been previously associated with SA. Hippocampal volumes were greater in SA compared to TOA (p1⁄40.03). Hippocampal volume, controlling for head size and age, was positively correlated with CVLT LDFR across both groups (r1⁄40.32, p1⁄40.001). There was a negative correlation between age and PiB in SA (r1⁄4-0.42, p1⁄40.035) that was not observed in TOA (r1⁄40.10, p1⁄40.39), such that only the younger SA participants had high global PiB burden. Conclusions: Our findings are consistent with previous work suggesting that SA are relatively spared from normal age-related cortical thinning and hippocampal volume loss. The correlation between hippocampal volume and memory performance indicates that preserved integrity of this structure plays a role in the maintenance of superior memory in SA. The novel observation that global PiB is inversely related to age in SA suggests that significant b-amyloid accumulation is not compatible with the ‘super’ aging trajectory.

Machine learning as a means toward precision diagnostics and prognostics

Machine learning plays an essential role in medical imaging. Pattern analysis techniques can identify, and quantify, subtle and spatially complex patterns of disease-induced changes in the brain despite confounding statistical noise and inter-individual variability. This allows the construction of sensitive biomarkers that can identify disease, or risk of developing it, and characterize future clinical progression on an individual patient basis. Thus pattern analysis techniques have become an indispensable part of the growing need for personalized, predictive medicine. However, despite important advances, several challenges remain before they can gain widespread acceptance as tools for precision diagnostics and prognostics in clinical practice. These include: (i) feature extraction and dimensionality reduction; (ii) readily interpreting complex multivariate models; and (iii) elucidating disease heterogeneity. In this chapter, we describe these challenges, putting emphasis on possible solutions, and present evidence of the usefulness of machine learning techniques at the clinical and research levels.