Showing papers by "Michael W. Vannier published in 2003"

Blind deblurring of spiral CT images

Abstract: To discriminate fine anatomical features in the inner ear, it has been desirable that spiral computed tomography (CT) may perform beyond their current resolution limits with the aid of digital image processing techniques. In this paper, we develop a blind deblurring approach to enhance image resolution retrospectively without complete knowledge of the underlying point spread function (PSF). An oblique CT image can be approximated as the convolution of an isotropic Gaussian PSF and the actual cross section. Practically, the parameter of the PSF is often unavailable. Hence, estimation of the parameter for the underlying PSF is crucially important for blind image deblurring. Based on the iterative deblurring theory, we formulate an edge-to-noise ratio (ENR) to characterize the image quality change due to deblurring. Our blind deblurring algorithm estimates the parameter of the PSF by maximizing the ENR, and deblurs images. In the phantom studies, the blind deblurring algorithm reduces image blurring by about 24%, according to our blurring residual measure. Also, the blind deblurring algorithm works well in patient studies. After fully automatic blind deblurring, the conspicuity of the submillimeter features of the cochlea is substantially improved.

Craniofacial computed tomography scanning: technology, applications and future trends.

Abstract: Structured Abstract Author – Vannier MW Objectives – Craniofacial imaging in three dimensions depends on computed tomography (CT) and related technologies. This paper explains the state-of-the-art for medical and dedicated craniofacial cone-beam CT scanners. Method – Current medical CT scanners are surveyed, especially the recently announced 16 simultaneous slice models with subsecond source–detector rotation times and spiral/helical third generation geometry. The medical scanner technology is contrasted with dedicated low-cost craniofacial cone-beam CT scanners to delineate the relevant technologies and clarify the differences. Results – CT scanners performance in any task is determined by their detectors and reconstruction algorithm primarily and to a lesser extent by the X-ray source, dose utilization, computational and display electronics, and software for post-processing. Each of these components differs between medical and low-cost cone-beam scanners, and the differences are tabulated and explained. Conclusion – Low-cost craniofacial CT scanners are significantly different from general purpose medical CT scanners, with compromises in technical performance. Despite their limitations, these instruments are remarkably useful for their intended application domain and should improve as computers continue to increase their performance.

Analytic modeling of breast elastography.

Abstract: The elastic moduli of tumors change during their pathological evolution. Elastographic imaging has potential for detecting and characterizing cancers by mapping the stiffness distribution in tissues. In this paper a micromechanics-based analytical method was developed to detect the location, size, and elastic modulus of a tumor mass embedded in a symmetric two-dimensional breast tissue. A closed-form solution for the strain elastograms (forward problem) was derived. A computational algorithm for the inverse problem was developed for the detection, localization, and characterization of a heterogeneous mass embedded in a breast tissue. Numerical examples were presented to evaluate the proposed method’s performance. The detectability of a tumor mass was estimated with respect to lesion location, size, and modulus contrast ratio. It was shown that the micromechanics theory provides a powerful tool for the diagnosis of breast cancer.

Craniofacial imaging informatics and technology development

Abstract: Structured Abstract Author– Vannier MW Purpose – ‘Craniofacial imaging informatics’ refers to image and related scientific data from the dentomaxillofacial complex, and application of ‘informatics techniques’ (derived from disciplines such as applied mathematics, computer science and statistics) to understand and organize the information associated with the data. Method – Major trends in information technology determine the progress made in craniofacial imaging and informatics. These trends include industry consolidation, disruptive technologies, Moore's law, electronic atlases and on-line databases. Each of these trends is explained and documented, relative to their influence on craniofacial imaging. Results – Craniofacial imaging is influenced by major trends that affect all medical imaging and related informatics applications. The introduction of cone beam craniofacial computed tomography scanners is an example of a disruptive technology entering the field. An important opportunity lies in the integration of biologic knowledge repositories with craniofacial images. Conclusion – The progress of craniofacial imaging will continue subject to limitations imposed by the underlying technologies, especially imaging informatics. Disruptive technologies will play a major role in the evolution of this field.

Characterizing Shape Differences between Phantom Image Populations via Multivariate Statistical Analysis of Inverse Consistent Transformations

Abstract: This paper presents an improved method for detecting statistically significant shape differences between image populations based on the multivariate Hotelling’s T2 test applied directly to image transformations. Performance of the method was evaluated using two phantom populations each consisting of 30 2D images with known average shape and known shape variability. Inverse-consistent linear-elastic image registration (ICLEIR) was used to construct a deformable template average image for both populations. The average image for the “normal” phantom population was used as the reference coordinate system to estimate ICLEIR correspondence transformations from the reference image to each population image. Following the work of Thirion et al., a multivariate two population Hotelling’s T2 test was performed on the displacement fields of these transformations at each voxel location in the reference coordinate system. We show that adding a conditioning constant e to the singular values of the sample covariance matrices used in the Hotelling’s T2 test reduces the false-positive rate. Furthermore, it is shown that adjusting the value of e focuses the statistical test response to the region of known shape differences present in the phantom image populations. Although limited, the phantom results presented in this paper provide baseline information for interpreting future results generated from real 3D medical images.

A study on the section sensitivity profile in multi-row-detector spiral CT.

Abstract: The section sensitivity profile (SSP) was well understood in the case of single-row-detector spiral CT. With the introduction of multi-row-detector spiral CT and the transition into cone-beam spiral CT, a revisit to the SSP issue becomes necessary. In this paper, the SSP of multi-row-detector spiral CT is formulated for the half-scan interpolation method at any transverse position. Based on the SSP formula, numerical simulation is performed to quantify the characteristics of the SSP with the number of detector rows up to 40. It is shown that the SSP varies as a function of the pitch and the number of detector rows. Given an appropriate selection of the pitch and the number of detector rows, the SSP does not change very much over the field of view in terms of the mean, the slice thickness, and the skewness of the SSP. Although in general applications the SSP at the gantry iso-center can be used as the representative of the SSP family, for more accurate analyses the spatial variation of the SSP must be taken into account.

Matching clinical and biological needs with emerging imaging technologies

Abstract: Acquisition, analysis, and integration of medical images with nonimage biomedical information are essential for research and clinical practice. Individualization of health care, assessment of disease risk, tailoring of therapies, and lifelong monitoring places extraordinary demands on emerging technologies in medical imaging, information management, and molecular medicine. To achieve the anticipated benefits in outcomes that new technology can deliver, imaging informatics infrastructure and tools must be available to represent, store, integrate, and present the data in a useful manner. Currently, images are not well integrated with other forms of biological knowledge. This paper outlines the clinical and research requirements for imaging, and identifies barriers to progress and links these needs with the current and future informatics technologies that may overcome them.

Inverse Consistent Image Registration of MR Brain Scans: Handedness in Normal Adult Males

Abstract: Brain shape differences between right and left-handed normal adults were evaluated by inverse-consistent linear-elastic image registration (ICLEIR) applied to MRI scans from two groups The study populations were 9 right-handed and 9 left-handed adult males from ages of 24 to 51 years old The mean brain shape of each population was computed and used as the reference shape for detecting shape differences Nonrigid, ICLEIR transformations that registered the mean brain image with the brain images from the pooled populations were used to detect local brain population shape differences Following the approach of Thirion et al, asymmetry maps between the left and right hemispheres of a brain image were computed by registering each brain image with their mirror images Local statistical shape differences between the two populations were determined using one and two-tailed t-tests at each voxel in the coordinate system of the mean brain shape Four t-tests were computed and compared which included the log-Jacobian and magnitude-divergence of the individual-to-pooled-average (IPA) correspondence map and the log-Jacobian and magnitude-divergence of the asymmetry maps Local shape differences between populations were evaluated to determine the location of asymmetries due to handedness Statistically significant (α=001) shape differences were found in this small pilot study with a sample size of 9 for each group Although the populations were too small to draw conclusions regarding neuromorphological differences between left and right handed individuals, the method shows promise for detecting brain shape differences between different populations

Automatic Measurement of the Labyrinth Using Image Registration and a Deformable Inner

Abstract: Materials and Methods. Computed tomography images of the inner ear are analyzed by placing them into a common orientation and then registering a digital atlas of the inner ear to the data set. The atlas is deformed from its original shape to match the shape of the inner ear in the computed tomography data set using inverse consistent elastic image registration. This process produces an individualized inner ear atlas containing subject-specific measurements and segmentations of the inner ear anatomy in the target computed tomography data set. The shape measurements include the volume and length of the cochlea, vestibule, semi-circular canals, and internal auditory canal; and the angles between the semi-circular canals. Results. A simulated population of inner ear shapes were generated based on the shape of a real population of inner ear shapes and were used to characterize the measurement error of this method. The deformable atlas was used to measure the shape of the left and right inner ear of six individuals. Conclusion. Measurement error for 15 of the 24 measurements of our simulated population had an average error of less than 1% and only one measurement had an average error greater than 2.54%. The deformable human inner ear atlas shows promise as a new method for automatically measuring the shape of the labyrinth. © AUR, 2003