Clustering High-Dimensional Landmark-Based Two-Dimensional Shape Data
Year: 2015
Type: article
Abstract: An important goal in image analysis is to cluster and recognize objects of interest according to the shapes of their boundaries. Clustering such objects faces at least four major challenges including a curved shape space, a high-dimensional feature space, a complex spatial correlation structure, and shape variation associated with some covariates (e.g., age or gender). The aim of this article is to develop a penalized model-based clustering framework to cluster landmark-based planar shape data, while explicitly addressing these challenges. Specifically, a mixture of offset-normal shape factor analyzers (MOSFA) is proposed with mixing proportions defined through a regression model (e.g., logistic) and an offset-normal shape distribution in each component for data in the curved shape space. A latent factor analysis model is introduced to explicitly model the complex spatial correlation. A penalized likelihood approach with both adaptive pairwise fused Lasso penalty function and L2 penalty function is used to automatically realize variable selection via thresholding and deliver a sparse solution. Our real data analysis has confirmed the excellent finite-sample performance of MOSFA in revealing meaningful clusters in the corpus callosum shape data obtained from the Attention Deficit Hyperactivity Disorder-200 (ADHD-200) study. Supplementary materials for this article are available online.
Authors: Chao Huang, Martin Styner, Hongtu Zhu
Institution: University of North Carolina at Chapel Hill
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: G1006888324, Biostatistics and Mental Health Neuroimaging and Genomics Training Grant, SCH: Proactive Health Monitoring Using Individualized Analysis of Tissue Elasticity, North Carolina Translational & Clinical Sciences Institute (NC TraCS), Developing Dynamic Tools for Analyzing Irregularly Spaced Longitudinal Affect Data, G4233579262, Advanced Statistical Tools for Ultra-High Dimensional Functional Data with Spatial-Temporal Correlation, G5573525031, Genome-wide Identification of Variants Affecting Early Human Brain Development, Statistical Analysis of Biomedical Imaging Data in Curved Space, UNC Clinical and Translational Science Award
OpenAlex ID: W1963539766
