Hamed Sari-Sarraf

A new X-ray computed tomography system for laboratory mouse imaging

Two versions of a new high-resolution X-ray computed tomography system are being developed to screen mutagenized mice in the Oak Ridge National Laboratory Mammalian Genetics Research Facility. The first prototype employs a single-pixel CdZnTe detector with a pinhole collimator operating in pulse counting mode. The second version employs a phosphor screen/CCD detector operating in current mode. The major system hardware includes a low-energy X-ray tube, two linear translation stages and a rotational stage. For the single-pixel detector, image resolution is determined by the step size of the detector stage; preliminary images have been acquired at 100 /spl mu/m and 250 /spl mu/m resolutions. The resolution of the phosphor screen detector is determined by the modulation transfer function of the phosphor screen; images with resolutions approaching 50 /spl mu/m have been acquired. The system performance with the two detectors is described and recent images are presented.

Hierarchical segmentation of cervical and lumbar vertebrae using a customized generalized Hough transform and extensions to active appearance models

The paper describes a semi-automatic segmentation method for application to cervical and lumbar X-ray images. The method consists of a three stage, coarse to fine, segmentation process utilizing the generalised Hough transform for one stage, and active appearance models for two stages. Customizations to these algorithms are introduced, and segmentation results for 273 cervical X-ray images and 262 lumbar X-ray images are presented.

A shift-invariant discrete wavelet transform

This article presents a unifying approach to the derivation and implementation of a shift-invariant wavelet transform of one- and two-dimensional (1-D and 2-D) discrete signals. Starting with Mallats (1989) multiresolution wavelet representation (MRWAR), it presents an analytical process through which a shift-invariant, orthogonal, discrete wavelet transform called the multiscale wavelet representation (MSWAR) is obtained. The coefficients in the MSWAR are shown to be inclusive of those in the MRWAR with the implication that the derived representation is invertible. The computational complexity of the MSWAR is quantified in terms of the required convolutions, and its implementation is shown to be equivalent to the filter upsampling technique.

Reconstruction of multi-energy X-ray computed tomography images of laboratory mice

A new X-ray computed tomography (CT) system is being developed at Oak Ridge National Laboratory to image laboratory mice for the purpose of rapid phenotype screening and identification. One implementation of this CT system allows simultaneous capture of several sets of sinogram data, each having a unique X-ray energy distribution. The goals of this paper are to (1) identify issues associated with the reconstruction of this energy-dependent data and (2) suggest preliminary approaches to address these issues. Due to varying numbers of photon counts within each set, both traditional (filtered backprojection, or FBP) and statistical (maximum likelihood, or ML) tomographic image reconstruction techniques have been applied to the energy-dependent sinogram data. Results of reconstructed images using both algorithms on sinogram data (high- and low-count) are presented. Also, tissue contrast within the energy-dependent images is compared to known X-ray attenuation coefficients of soft tissue (e.g. muscle, bone, and fat).

Customized Hough transform for robust segmentation of cervical vertebrae from X-ray images

This paper addresses the issues involved in developing a robust segmentation technique capable of finding the location and orientation of the cervical vertebrae in X-ray images. This technique should be invariant to rotation, scale, noise, occlusions and shape variability. A customized approach, based on the generalized Hough transform (GHT), that captures shape variability and exploits shape information embedded in the accumulator structure to overcome noise and occlusions is proposed. This approach effectively finds estimates of the location and orientation of the cervical vertebrae boundaries in digitized X-ray images.

Robust defect segmentation in woven fabrics

This paper describes a robust segmentation algorithm for the detection and localization of woven fabric defects. The essence of the presented segmentation algorithm is the localization of those events (i.e., defects) in the input images that disrupt the global homogeneity of the background texture. To this end, preprocessing modules, based on the wavelet transform and edge fusion, are employed with the objective of attenuating the background texture and accentuating the defects. Then, texture features are utilized to measure the global homogeneity of the output images. If these images are deemed to be globally nonhomogeneous (i.e., defects are present), a local roughness measure is used to localize the defects. The utility of this algorithm can be extended beyond the specific application in our work, that is, defect segmentation in woven fabrics. Indeed, in a general sense, this algorithm can be used to detect and to localize anomalies that reside in images characterized by ordered texture. The efficacy of this algorithm has been tested thoroughly under realistic conditions and as a part of an on-line fabric inspection system. Using over 3700 images of fabrics, containing 26 different types of defects, the overall detection rate of our approach was 89% with a localization accuracy of less than 0.2 inches and a false alarm rate of 2.5%.

A new deformable model for analysis of X-ray CT images in preclinical studies of mice for polycystic kidney disease

This paper describes the application of a new probabilistic shape and appearance model (PSAM) algorithm to the task of detecting polycystic kidney disease (PKD) in X-ray computed tomography images of laboratory mice. The genetically engineered PKD mouse is a valuable animal model that can be used to develop new treatments for kidney-related problems in humans. PSAM is a statistical-based deformable model that improves upon existing point distribution models for boundary-based object segmentation. This new deformable model algorithm finds the optimal boundary position using an objective function that has several unique characteristics. Most importantly, the objective function includes both global shape and local gray-level characteristics, so optimization occurs with respect to both pieces of information simultaneously. PSAM is employed to segment the mouse kidneys and then texture measurements are applied within kidney boundaries to detect PKD. The challenges associated with the segmentation nonrigid organs along with the availability of a priori information led to the choice of a trainable, deformable model for this application. In 103 kidney images that were analyzed as part of a preclinical animal study, the mouse kidneys and spine were segmented with an average error of 2.4 pixels per boundary point. In all 103 cases, the kidneys were successfully segmented at a level where PKD could be detected using mean-of-local-variance texture measurements within the located boundary.

Online optical measurement and monitoring of yarn density in woven fabrics

This paper describes a vision-based system that monitors the yarn density of woven fabrics on-line. The system is described in terms of its two principal modules, namely, the image acquisition and the image analysis subsystems. The image acquisition subsystem is implemented with standard components on a low-cost personal computer platform. These components consist of a line-scan camera, a DSP-based image acquisition and processing card, and a host personal computer. The image acquisition process is controlled by a software module that runs on the DSP board and accumulates a 2D image suitable for the density measurement algorithm. The image analysis subsystem, which also runs on the DSP board, implements a novel, yet straightforward, algorithm that utilizes the discrete Fourier transform for monitoring the yarn density of the fabrics from the acquired images. In this algorithm, the Fourier spectrum of the images is covered by contiguous, concentric annular regions that have a prespecified width. THe spectrum values within each annular region are summed, normalized, and subsequently used to produce a 1D signature. Simple statistics of the obtained signatures are the basis for characterizing the fabric in terms of its yarn density. The described system is tested on seven fabrics with common properties but varying yarn densities and has shown to be accurate within 2 yarns per inch in either direction. It is also shown that the obtained accuracy, which is primarily a function of the image resolution, can be greatly improved.

Volumetric segmentation via 3D active shape models

A volumetric image segmentation algorithm has been developed and implemented by extending a 2D algorithm based on active shape models. The new technique allows segmentation of 3D objects that are embedded within volumetric image data. The extension from 2D involved four components: landmarking, shape modeling, gray-level modeling, and segmentation. Algorithms and software tools have been implemented to allow a user to efficiently landmark a 3D object training set. Additional tools were built that subsequently,, generate models of 3D object shape and gray-level appearance based on this training data. An object segmentation strategy was implemented that optimizes these models to segment a previously unseen instance of the object. Results of this new 3D segmentation algorithm have been generated for a synthetic volumetric data set.

A novel approach to computer-aided diagnosis of mammographic images

The article is a work-in-progress report of a research endeavor that deals with the design and development of a novel approach to computer aided diagnosis (CAD) of mammographic images. With the initial emphasis being on the analysis of microcalcifications, the proposed approach defines a synergistic paradigm that utilizes new methodologies together with previously developed techniques. The new paradigm is intended to promote a higher degree of accuracy in CAD of mammograms with an increased overall throughput. The process of accomplishing these goals is initiated by the fractal encoding of the input image, which gives rise to the generation of focus-of-attention regions (FARs), that is, regions that contain anomalies. The primary thrust of this work is to demonstrate that by considering FARs, rather than the entire input image, the performances of the ensuing processes (i.e., segmentation, feature extraction, and classification) are enhanced in terms of accuracy and speed. An experimental study is included that demonstrates the impact of FAR generation on the process of microcalcification segmentation.