Ingmar Bitter

Penalized-distance volumetric skeleton algorithm

Introduces a refined general definition of a skeleton that is based on a penalized distance function and that cannot create any of the degenerate cases of the earlier CEASAR (Center-line Extraction Algorithm-Smooth, Accurate and Robust) and TEASAR (Tree-structure Extraction Algorithm for Skeletons-Accurate and Robust) algorithms. Additionally, we provide an algorithm that finds the skeleton accurately and rapidly. Our solution is fully automatic, which frees the user from having to engage in manual data pre-processing. We present the accurate skeletons computed on a number of test data sets. The algorithm is very efficient, as demonstrated by the running times, which were all below seven minutes.

Automatic centerline extraction for virtual colonoscopy

In this paper, we introduce a concise and concrete definition of an accurate colon centerline and provide an efficient automatic means to extract the centerline and its associated branches (caused by a forceful touching of colon and small bowel or a deep fold in twisted colon lumen). We further discuss its applications on fly-through path planning and endoscopic simulation, as well as its potential to solve the challenging touching and colon collapse problems in virtual colonoscopy. Experimental results demonstrated its centeredness, robustness, and efficiency.

Dynamics of phosphatidylinositol 4,5-bisphosphate in actin-rich structures

Phosphatidylinositol 4,5-bisphosphate (PI(4,5)P(2)) is known to regulate a wide range of molecular targets and cellular processes, from ion channels to actin polymerization [1] [2] [3] [4] [5] [6]. Recent studies have used the phospholipase C-delta1 (PLC-delta1) pleckstrin-homology (PH) domain fused to green fluorescent protein (GFP) as a detector for PI(4,5)P(2) in vivo [7] [8] [9] [10]. Although these studies demonstrated that PI(4,5)P(2) is concentrated in the plasma membrane, its association with actin-containing structures was not reported. In the present study, fluorescence imaging of living NIH-3T3 fibroblasts expressing the PLC-delta1 PH domain linked to enhanced green fluorescent protein (PH-EGFP) reveals intense, non-uniform fluorescence in distinct structures at the cell periphery. Corresponding fluorescence and phase-contrast imaging over time shows that these fluorescent structures correlate with dynamic, phase-dense features identified as ruffles and with microvillus-like protrusions from the cells dorsal surface. Imaging of fixed and permeabilized cells shows co-localization of PH-EGFP with F-actin in ruffles, but not with vinculin in focal adhesions. The selective concentration of the PH-EGFP fusion protein in highly dynamic regions of the plasma membrane that are rich in F-actin supports the hypothesis that localized synthesis and lateral segregation of PI(4,5)P(2) spatially restricts actin polymerization and thereby affects cell spreading and retraction.

High-quality volume rendering using texture mapping hardware

Wt present a method Jor volume rendering of regular grids cclhic~h takes advantage of <?D texture mapping hardware currc,rhlly available on graphics workstations. Our method products accurate shadang for arbitrary and dynamically changing directionul lights, viewing parameters, and transfer funclior~. TIlis is achieved by hardware interpolating the data values and gradients before software classification and shadrng. The method works equally well for parallel and perspective projections. We present two approaches for OUT method: one which takes advantage of software ray casting optimitaIrons nnd another which takes advantage of hardware blending (Acceleration. CR Categories: 13.1 [Computer Graphics]: Hardware Architecture; 1.3.3 [Computer Graphics]: Picture/Image Generation; 1.3.7 [Computer Graphics]: Three-Dimensional Graphics and Realism-Color, shading, shadowing, and texture

CEASAR: a smooth, accurate and robust centerline extraction algorithm

We present CEASAR, a centerline extraction algorithm that delivers smooth, accurate and robust results. Centerlines are needed for accurate measurements of length along winding tubular structures. Centerlines are also required in automatic virtual navigation through human organs, such as the colon or the aorta, as they are used to control movement and orientation of the virtual camera. We introduce a concise but general definition of a centerline, and provide an algorithm that finds the centerline accurately and rapidly. Our algorithm is provably correct for general geometries. Our solution is fully automatic, which frees the user from having to engage in data preprocessing. For a number of test datasets, we show the smooth and accurate centerlines computed by our CEASAR algorithm on a single 194 MHz MIPS R10000 CPU within five minutes.

TEASAR: tree-structure extraction algorithm for accurate and robust skeletons

We introduce the TEASAR algorithm which is a treestructure extraction algorithm delivering skeletons that are accurate and robust. Volumetric skeletons are needed for accurate measurements of length along branching and winding structures. Skeletons are also required in automatic virtual navigation, such as traveling through human organs (e.g., the colon) to control movement and orientation of the virtual camera. We introduce a concise but general definition of a skeleton, and provide an algorithm that finds the skeleton accurately and rapidly. Our solution is fully automatic, which frees the user from having to engage in data preprocessing. We present the accurate skeletons computed on a number of test datasets. The algorithm is efficient as demonstrated by the running times on a single 194 MHz MIPS R10000 CPU which were all below five minutes.

Image segmentation approach to extract colon lumen through colonic material tagging and hidden Markov random field model for virtual colonoscopy

Virtual colonoscopy provides a safe, minimal-invasive approach to detect colonic polyps using medical imaging and computer graphics technologies. Residual stool and fluid are problematic for optimal viewing of the colonic mucosa. Electronic cleansing techniques combining bowel preparation, oral contrast agents, and image segmentation were developed to extract the colon lumen from computed tomography (CT) images of the colon. In this paper, we present a new electronic colon cleansing technology, which employs a hidden Markov random filed (MRF) model to integrate the neighborhood information for overcoming the non-uniformity problems within the tagged stool/fluid region. Prior to obtaining CT images, the patient undergoes a bowel preparation. A statistical method for maximum a posterior probability (MAP) was developed to identify the enhanced regions of residual stool/fluid. The method utilizes a hidden MRF Gibbs model to integrate the spatial information into the Expectation Maximization (EM) model-fitting MAP algorithm. The algorithm estimates the model parameters and segments the voxels iteratively in an interleaved manner, converging to a solution where the model parameters and voxel labels are stabilized within a specified criterion. Experimental results are promising.

A ray-slice-sweep volume rendering engine

Ray-slice-sweeping is a plane sweep algorithm for volume rendering, The compositing buffer sweeps through the volume and combines the accumulated image with the new slice of just-projected voxels. The image combination is guided by sight rays from the view point through every voxel of the new slice. Cube-.#L is a volume rendering architecture which employs a ray-slice-sweeping algorithm. It improves the Cube-4 architecture in three ways. First, during perspective projection all voxels of the dataset contribute to the rendering. Second, it computes gradients at the voxel positions which improves accuracy and allows a more compact implementation, Third, Cube-AL has less control overhead than Cube-C

Adaptive perspective ray casting

We present a method to accurately and efficiently perform perspective volumetric ray casting of uniform regular datasets, called Exponential-Region (ER) Perspective. Unlike previous methods which undersample, oversample, or approximate the data, our method near uniformly samples the data throughout the viewing volume. In addition, it gains algorithmic advantages from a regular sampling pattern and cache-coherent read access, making it an algorithm well suited for implementation on hardware architectures for volume rendering. We qualify the algorithm by its filtering characteristics and demonstrate its effectiveness by contrasting its antialiasing quality and timing with other perspective ray casting methods.

3D virtual colonoscopy with real-time volume rendering

In our previous work, we developed a virtual colonoscopy system on a high-end 16-processor SGI Challenge with an expensive hardware graphics accelerator. The goal of this work is to port the system to a low cost PC in order to increase its availability for mass screening. Recently, Mitsubishi Electric has developed a volume-rendering PC board, called VolumePro, which includes 128 MB of RAM and vg500 rendering chip. The vg500 chip, based on Cube-4 technology, can render a 2563 volume at 30 frames per second. High image quality of volume rendering inside the colon is guaranteed by the full lighting model and 3D interpolation supported by the vg500 chip. However, the VolumePro board is lacking some features required by our interactive colon navigation. First, VolumePro currently does not support perspective projection which is paramount for interior colon navigation. Second, the patient colon data is usually much larger than 2563 and cannot be rendered in real-time. In this paper, we present our solutions to these problems, including simulated perspective projection and axis aligned boxing techniques, and demonstrate the high performance of our virtual colonoscopy system on low cost PCs.