Joon Kyung Seong

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 Alzheimers 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 Alzheimers 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.

Trimming local and global self-intersections in offset curves/surfaces using distance maps

A robust and efficient algorithm for trimming both local and global self-intersections in offset curves and surfaces is presented. Our scheme is based on the derivation of a rational distance map between the original curve or surface and its offset. By solving a bivariate polynomial equation for an offset curve or a system of three polynomial equations for an offset surface, all local and global self-intersection regions in offset curves or surfaces can be detected. The zero-set of polynomial equation(s) corresponds to the self-intersection regions. These regions are trimmed by projecting the zero-set into an appropriate parameter space. The projection operation simplifies the analysis of the zero-set, which makes the proposed algorithm numerically stable and efficient. Furthermore, in a post-processing step, a numerical marching method is employed, which provides a highly precise scheme for self-intersection elimination in both offset curves and surfaces. The effectiveness of our approach is demonstrated using several experimental results.

Sweep-based human deformation

We present a sweep-based approach to human body modeling and deformation. A rigid 3D human model, given as a polygonal mesh, is approximated with control sweep surfaces. The vertices on the mesh are bound to nearby sweep surfaces and then follow the deformation of the sweep surfaces as the model bends and twists its arms, legs, spine and neck. Anatomical features including bone-protrusion, muscle-bulge, and skin-folding are supported by a GPU-based collision detection procedure. The volumes of arms, legs, and torso are kept constant by a simple control using a volume integral formula for sweep surfaces. We demonstrate the effectiveness of this sweep-based human deformation in several test animation clips.

Changes in subcortical structures in early- versus late-onset Alzheimer's disease

Patients with early-onset Alzheimers disease (EOAD) are reported to be different from those with late-onset Alzheimers disease (LOAD) in terms of neuropsychological and neuroimaging findings. In this study, we aimed to compare the longitudinal volume changes of 6 subcortical structures (the amygdala, hippocampus, thalamus, putamen, globus pallidus, and caudate nucleus) between patients with EOAD and LOAD for 3 years. We prospectively recruited 36 patients with probable Alzheimers disease (14 EOAD, 22 LOAD) and 14 normal control subjects. We analyzed the volume of subcortical structures using an automatic surface-based method. At baseline, there were no differences in the volumes of subcortical structures between patients with EOAD and LOAD. However, over 3 years of longitudinal follow-up, patients with EOAD showed more rapid volumetric decline in the caudate, putamen, and thalamus than patients with LOAD, which is consistent with neuropsychological results. Our findings suggested that the cognitive reserve theory might be applicable to explain different decline rates of the volumes of the basal ganglia and thalamus according to onset age.

Quantitative comparison and analysis of sulcal patterns using sulcal graph matching: A twin study

The global pattern of cortical sulci provides important information on brain development and functional compartmentalization. Sulcal patterns are routinely used to determine fetal brain health and detect cerebral malformations. We present a quantitative method for automatically comparing and analyzing the sulcal pattern between individuals using a graph matching approach. White matter surfaces were reconstructed from volumetric T1 MRI data and sulcal pits, the deepest points in local sulci, were identified on this surface. The sulcal pattern was then represented as a graph structure with sulcal pits as nodes. The similarity between graphs was computed with a spectral-based matching algorithm by using the geometric features of nodes (3D position, depth and area) and their relationship. In particular, we exploited the feature of graph topology (the number of edges and the paths between nodes) to highlight the interrelated arrangement and patterning of sulcal folds. We applied this methodology to 48 monozygotic twins and showed that the similarity of the sulcal graphs in twin pairs was significantly higher than in unrelated pairs for all hemispheres and lobar regions, consistent with a genetic influence on sulcal patterning. This novel approach has the potential to provide a quantitative and reliable means to compare sulcal patterns.

A multi-resolution scheme for distortion-minimizing mapping between human subcortical structures based on geodesic construction on Riemannian manifolds

In this paper, we deal with a subcortical surface registration problem. Subcortical structures including hippocampi and caudates have a small number of salient features such as heads and tails unlike cortical surfaces. Therefore, it is hard, if not impossible, to perform subcortical surface registration with only such features. It is also non-trivial for neuroanatomical experts to select landmarks consistently for subcortical surfaces of different subjects. We therefore present a landmark-free approach for subcortical surface registration by measuring the amount of mesh distortion between subcortical surfaces assuming that the surfaces are represented by meshes. The input meshes can be constructed using any surface modeling tool available in the public domain since our registration method is independent of a surface modeling process. Given the source and target surfaces together with their representing meshes, the vertex positions of the source mesh are iteratively displaced while preserving the underlying surface shape in order to minimize the distortion to the target mesh. By representing each surface mesh as a point on a high-dimensional Riemannian manifold, we define a distance metric on the manifold that measures the amount of distortion from a given source mesh to the target mesh, based on the notion of isometry while penalizing triangle flipping. Under this metric, we reduce the distortion minimization problem to the problem of constructing a geodesic curve from the moving source point to the fixed target point on the manifold while satisfying the shape-preserving constraint. We adopt a multi-resolution framework to solve the problem for distortion-minimizing mapping between the source and target meshes. We validate our registration scheme through several experiments: distance metric comparison, visual validation using real data, robustness test to mesh variations, feature alignment using anatomic landmarks, consistency with previous clinical findings, and comparison with a surface-based registration method, LDDMM-surface.

Amyloid burden, cerebrovascular disease, brain atrophy, and cognition in cognitively impaired patients

We investigated the independent effects of Alzheimers disease (AD) and cerebrovascular disease (CVD) pathologies on brain structural changes and cognition.

Shape Changes of the Basal Ganglia and Thalamus in Alzheimer's Disease: A Three-Year Longitudinal Study

BACKGROUND A large number of Alzheimers disease (AD) studies have focused on medial temporal and cortical atrophy, while changes in the basal ganglia or thalamus have received less attention. OBJECTIVE The aim of this study was to investigate the existence of progressive topographical shape changes in the basal ganglia (caudate nucleus, putamen, and globus pallidus) and thalamus concurrent with AD disease progression over three years. This study also examined whether declines in volumes of the basal ganglia or thalamus might be responsible for cognitive decline in patients with AD. METHODS Thirty-six patients with early stage AD and 14 normal control subjects were prospectively recruited for this study. All subjects were assessed with neuropsychological tests and MRI at baseline and Years 1 and 3. A longitudinal shape analysis of the basal ganglia and thalamus was performed by employing a boundary surface-based shape analysis method. RESULTS AD patients exhibited specific regional atrophy in the right caudate nucleus and the bilateral putamen at baseline, and as the disease progressed, regional atrophic changes in the left caudate nucleus were found to conform to a distinct topography after controlling the total brain volume. Volumetric decline of the caudate nucleus and putamen correlated with cognitive decline in frontal function after controlling for age, gender, education, follow-up years, and total brain volume changes. CONCLUSION Our findings suggest that shape changes of the basal ganglia occurred regardless of whole brain atrophy as AD progressed and were also responsible for cognitive decline that was observed from the frontal function tests.

A Network Flow-based Analysis of Cognitive Reserve in Normal Ageing and Alzheimer’s Disease

Cognitive reserve is the ability to sustain cognitive function even with a certain amount of brain damages. Here we investigate the neural compensation mechanism of cognitive reserve from the perspective of structural brain connectivity. Our goal was to show that normal people with high education levels (i.e., cognitive reserve) maintain abundant pathways connecting any two brain regions, providing better compensation or resilience after brain damage. Accordingly, patients with high education levels show more deterioration in structural brain connectivity than those with low education levels before symptoms of Alzheimer’s disease (AD) become apparent. To test this hypothesis, we use network flow measuring the number of alternative paths between two brain regions in the brain network. The experimental results show that for normal aging, education strengthens network reliability, as measured through flow values, in a subnetwork centered at the supramarginal gyrus. For AD, a subnetwork centered at the left middle frontal gyrus shows a negative correlation between flow and education, which implies more collapse in structural brain connectivity for highly educated patients. We conclude that cognitive reserve may come from the ability of network reorganization to secure the information flow within the brain network, therefore making it more resistant to disease progress.

Hippocampal volume and shape in pure subcortical vascular dementia.

The purposes of the present study were to explore whether hippocampal atrophy exists in pure subcortical vascular dementia (SVaD) as defined by negative (11)C-Pittsburg compound-B (PiB(-)) positron emission tomography and to compare hippocampal volume and shape between PiB(-) SVaD and PiB positive (PiB(+)) Alzheimers disease (AD) dementia. Hippocampal volume and shape were compared among 40 patients with PiB(-) SVaD, 34 with PiB(+) AD, and 21 elderly with normal cognitive function (NC). The normalized hippocampal volume of PiB(-) SVaD was significantly smaller than NC but larger than that of PiB(+) AD (NC > PiB(-) SVaD > PiB(+) AD). Both PiB(-) SVaD and PiB(+) AD patients had deflated shape changes in the cornus ammonis (CA) 1 and subiculum compared with NC. However, direct comparison between PiB(-) SVaD and PiB(+) AD demonstrated more inward deformity in the subiculum of the left hippocampus in PiB(+) AD. PiB(-) SVaD patients did have smaller hippocampal volumes and inward shape change on CA 1 and subiculum compared with NC, suggesting that cumulative ischemia without amyloid pathology could lead to hippocampal atrophy and shape changes.