Ilwoo Lyu

Spectral-based automatic labeling and refining of human cortical sulcal curves using expert-provided examples

We present a spectral-based method for automatically labeling and refining major sulcal curves of a human cerebral cortex. Given a set of input (unlabeled) sulcal curves automatically extracted from a cortical surface and a collection of expert-provided examples (labeled sulcal curves), our objective is to identify the input major sulcal curves and assign their neuroanatomical labels, and then refines these curves based on the expert-provided example data, without employing any atlas-based registration scheme as preprocessing. In order to construct the example data, neuroanatomists manually labeled a set of 24 major sulcal curves (12 each for the left and right hemispheres) for each individual subject according to a precise protocol. We collected 30 sets of such curves from 30 subjects. Given the raw input sulcal curve set of a subject, we choose the most similar example curve to each input curve in the set to label and refine the latter according to the former. We adapt a spectral matching algorithm to choose the example curve by exploiting the sulcal curve features and their relationship. The high dimensionality of sulcal curve data in spectral matching is addressed by using their multi-resolution representations, which greatly reduces time and space complexities. Our method provides consistent labeling and refining results even under high variability of cortical sulci across the subjects. Through experiments we show that the results are comparable in accuracy to those done manually. Most output curves exhibited accuracy values higher than 80%, and the mean accuracy values of the curves in the left and the right hemispheres were 84.69% and 84.58%, respectively.

Group-wise cortical correspondence via sulcal curve-constrained entropy minimization

We present a novel cortical correspondence method employing group-wise registration in a spherical parametrization space for the use in local cortical thickness analysis in human and non-human primate neuroimaging studies. The proposed method is unbiased registration that estimates a continuous smooth deformation field into an unbiased average space via sulcal curve-constrained entropy minimization using spherical harmonic decomposition of the spherical deformation field. We initialize a correspondence by our pair-wise method that establishes a surface correspondence with a prior template. Since this pair-wise correspondence is biased to the choice of a template, we further improve the correspondence by employing unbiased ensemble entropy minimization across all surfaces, which yields a deformation field onto the iteratively updated unbiased average. The specific entropy metric incorporates two terms: the first focused on optimizing the correspondence of automatically extracted sulcal landmarks and the second on that of sulcal depth maps. We also propose an encoding scheme for spherical deformation via spherical harmonics as well as a novel method to choose an optimal spherical polar coordinate system for the most efficient deformation field estimation. The experimental results show evidence that the proposed method improves the correspondence quality in non-human primate and human subjects as compared to the pair-wise method.

Geodesic Distances to Landmarks for Dense Correspondence on Ensembles of Complex Shapes

Establishing correspondence points across a set of biomedical shapes is an important technology for a variety of applications that rely on statistical analysis of individual subjects and populations. The inherent complexity (e.g. cortical surface shapes) and variability (e.g. cardiac chambers) evident in many biomedical shapes introduce significant challenges in finding a useful set of dense correspondences. Application specific strategies, such as registration of simplified (e.g. inflated or smoothed) surfaces or relying on manually placed landmarks, provide some improvement but suffer from limitations including increased computational complexity and ambiguity in landmark placement. This paper proposes a method for dense point correspondence on shape ensembles using geodesic distances to a priori landmarks as features. A novel set of numerical techniques for fast computation of geodesic distances to point sets is used to extract these features. The proposed method minimizes the ensemble entropy based on these features, resulting in isometry invariant correspondences in a very general, flexible framework.

Automatic sulcal curve extraction on the human cortical surface

The recognition of sulcal regions on the cortical surface is an important task to shape analysis and landmark detection. However, it is challenging especially in a complex, rough human cortex. In this paper, we focus on the extraction of sulcal curves from the human cortical surface. The previous sulcal extraction methods are time-consuming in practice and often have a difficulty to delineate curves correctly along the sulcal regions in the presence of significant noise. Our pipeline is summarized in two main steps: 1) We extract candidate sulcal points spread over the sulcal regions. We further reduce the size of the candidate points by applying a line simplification method. 2) Since the candidate points are potentially located away from the exact valley regions, we propose a novel approach to connect candidate sulcal points so as to obtain a set of complete curves (line segments). We have shown in experiment that our method achieves high computational efficiency, improved robustness to noise, and high reliability in a test-retest situation as compared to a well-known existing method.

Novel Local Shape-Adaptive Gyrification Index with Application to Brain Development

Conventional approaches to quantification of the cortical folding employ a simple circular kernel. Such a kernel commonly covers multiple cortical gyral/sulcal regions that may be functionally unrelated and also often blurs local gyrification measurements. We propose a novel adaptive kernel for quantification of the local cortical folding, which incorporates neighboring gyral crowns and sulcal fundi. The proposed kernel is adaptively elongated to cover regions along the cortical folding patterns. The experimental results showed that the proposed kernel-based gyrification measure achieved a higher reproducibility in a multi-scan human phantom dataset and captured the cortical folding in a more shape-adaptive way than the conventional method. In early human brain development, we found positive correlations with age over most cortical regions as previously found as well as novel, refined regions of both positive and negative correlations undetectable by the conventional method.

Cortical surface shape assessment via sulcal/gyral curve-based gyrification index

We propose novel gyrification index (GI) based on sulcal and gyral curves on the human cortical surface. Instead of using the widely employed methods based on Euclidean or geodesic kernels of uniform size, we biologically determine local regions based on a curve-wise correspondence between sulcal and gyral curves. Specifically, we initially extract sulcal and gyral curves from the cortical surface. For each sulcal point, we then find two corresponding gyral points associated with the closest gyral curves. This process requires a categorical optimization that generally possesses an intractable parameter space, which is addressed via a novel evolutionary algorithm. Finally, we propagate sparse measurements of the proposed sulcal GI at each sulcal point to the entire cortical surface in order to yield a complete GI map. The experimental results show that our measurement achieves reasonable reliability across a scan-rescan dataset and provides a complementary information of cortical folding, compared with a recent kernel-based metric in a longitudinal study.

Cortical correspondence via sulcal curve-constrained spherical registration with application to Macaque studies

In this work, we present a novel cortical correspondence method with application to the macaque brain. The correspondence method is based on sulcal curve constraints on a spherical deformable registration using spherical harmonics to parameterize the spherical deformation. Starting from structural MR images, we first apply existing preprocessing steps: brain tissue segmentation using the Automatic Brain Classification tool (ABC), as well as cortical surface reconstruction and spherical parametrization of the cortical surface via Constrained Laplacian-based Automated Segmentation with Proximities (CLASP). Then, initial correspondence between two cortical surfaces is automatically determined by a curve labeling method using sulcal landmarks extracted along sulcal fundic regions. Since the initial correspondence is limited to sulcal regions, we use spherical harmonics to extrapolate and regularize this correspondence to the entire cortical surface. To further improve the correspondence, we compute a spherical registration that optimizes the spherical harmonic parameterized deformation using a metric that incorporates the error over the sulcal landmarks as well as the normalized cross correlation of sulcal depth maps over the whole cortical surface. For evaluation, a normal 18-months-old macaque brain (for both left and right hemispheres) was matched to a prior macaque brain template with 9 manually labeled, major sulcal curves. The results show successful registration using the proposed registration approach. Evaluation results for optimal parameter settings are presented as well.

Particle-Guided Image Registration

We present a novel image registration method based on B-spline free-form deformation that simultaneously optimizes particle correspondence and image similarity metrics. Different from previous B-spline based registration methods optimized w.r.t. the control points, the deformation in our method is estimated from a set of dense unstructured pair of points, which we refer as corresponding particles. As intensity values are matched on the corresponding location, the registration performance is iteratively improved. Moreover, the use of corresponding particles naturally extends our method to a group-wise registration by computing a mean of particles. Motivated by a surface-based group-wise particle correspondence method, we developed a novel system that takes such particles to the image domain, while keeping the spirit of the method similar. The core algorithm both minimizes an entropy based group-wise correspondence metric as well as maximizes the space sampling of the particles. We demonstrate the results of our method in an application of rodent brain structure segmentation and show that our method provides better accuracy in two structures compared to other registration methods.

Multi-atlas segmentation with particle-based group-wise image registration

We propose a novel multi-atlas segmentation method that employs a group-wise image registration method for the brain segmentation on rodent magnetic resonance (MR) images. The core element of the proposed segmentation is the use of a particle-guided image registration method that extends the concept of particle correspondence into the volumetric image domain. The registration method performs a group-wise image registration that simultaneously registers a set of images toward the space defined by the average of particles. The particle-guided image registration method is robust with low signal-to-noise ratio images as well as differing sizes and shapes observed in the developing rodent brain. Also, the use of an implicit common reference frame can prevent potential bias induced by the use of a single template in the segmentation process. We show that the use of a particle guided-image registration method can be naturally extended to a novel multi-atlas segmentation method and improves the registration method to explicitly use the provided template labels as an additional constraint. In the experiment, we show that our segmentation algorithm provides more accuracy with multi-atlas label fusion and stability against pair-wise image registration. The comparison with previous group-wise registration method is provided as well.

Multiple atlases-based joint labeling of human cortical sulcal curves

We present a spectral-based sulcal curve labeling method by considering geometrical information of neighboring curves in a multiple atlases-based framework. Compared to the conventional method, we propose to use neighboring curves for avoiding ambiguity in curve-by-curve labeling and to integrate the labeling results obtained from multiple atlases for consistent labeling. In particular, we compute a histogram of points on the neighboring curves as a new feature descriptor for each point on a sulcal curve under consideration. To better resolve ambiguity in the curve labeling, we also employ the neighboring curves that are parallel to major sulcal curves. Moreover, we further integrate all the results from multiple atlases into a linear system, by solving which our method ultimately gives accurate labels to the major curves in the subjects. Experimental results on evaluation of 12 major sulcal curves of 12 human cortical surfaces indicate that our method achieves higher labeling accuracy 7.87% compared to the conventional method, while reducing 4.41% of false positive labeling errors on average.