Intelligent Initialization and Interactivity: Optimizing Level Sets for T1-weighted White Matter Segmentation
نویسندگان
چکیده
White matter (WM) segmentation from T1-weighted MRI is complicated by intensity non-uniformities, noise, and WM’s high surface area to volume ratio. Accurate algorithms are often computationally intensive and time consuming, precluding interactivity and routine clinical use. To address this we developed a workand step-efficient parallel narrow-band level set algorithm and mapped this onto commodity GPU hardware. Our algorithm can segment brain WM in 3 seconds. However, it requires expert tuning of 3 parameters. Here we describe recent efforts to improve the precision, accuracy and simplicity of WM segmentation by: a) intelligently initializing algorithm parameters; and, b) allowing interactive parameter tuning during algorithm execution, along with real-time 2D and 3D visualization of parameter effects on segmentation results.
منابع مشابه
A Segmentation Algorithm for Brain Mr Images Using Fuzzy Model and Level Sets
This paper presents a novel algorithm based on level set techniques for tissue segmentation of brain magnetic resonance (MR) images. The method initially proposed by Suri is improved by using a new regional term based on the investigation and analysis of its stability. The improved algorithm solves the stability problem associated with the original algorithm resulting in a greatly improved qual...
متن کاملAutomated MRI brain tissue segmentation based on mean shift and fuzzy c-means using a priori tissue probability maps
This paper presents a novel fully automated unsupervised framework for the brain tissue segmentation in magnetic resonance (MR) images. The framework is a combination of Bayesian-based adaptive mean shift, a priori spatial tissue probability maps and fuzzy c-means. Mean shift is employed to cluster the tissues in the joint spatial-intensity feature space and then a fuzzy c-means is applied with...
متن کاملFully automatic brain segmentation using model-guided level sets and skeleton-based models
A fully automatic brain segmentation method is presented. First the skull is stripped using a model-based level set on T1-weighted inversion recovery images, then the brain ventricles and basal ganglia are segmented using the same method on T1-weighted images. The central white matter is segmented using a regular level set method but with high curvature regulation. To segment the cortical gray ...
متن کاملMulti-phase Three-Dimensional Level Set Segmentation of Brain MRI
This paper presents the implementation and quantitative evaluation of a four-phase three-dimensional active contour implemented with a level set framework for automated segmentation of cortical structures on brain T1 MRIs. The segmentation algorithm performed an optimal partitioning of threedimensional data based on homogeneity measures that naturally evolves to the extraction of different tiss...
متن کاملEmploying Visual Analytics to Aid the Design of White Matter Hyperintensity Classifiers
Accurate segmentation and quantification of brain white matter hyperintensities (WMHs) is important for prognosis and disease monitoring. Conventional WMH segmentation techniques employ classifiers, trained on T1 and FLAIR weighted MR images. Yet, there are indications that including features derived from diffusion weighted MRI, can improve classification. However, many features can be computed...
متن کامل