By Topic

Biomedical Image Analysis, 1994., Proceedings of the IEEE Workshop on

Date 24-25 June 1994

Filter Results

Displaying Results 1 - 25 of 35
  • Proceedings of IEEE Workshop on Biomedical Image Analysis

    Publication Year: 1994
    Save to Project icon | Request Permissions | PDF file iconPDF (21 KB)  
    Freely Available from IEEE
  • Brain image registration based on curve mapping

    Publication Year: 1994 , Page(s): 245 - 254
    Cited by:  Papers (12)  |  Patents (8)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1156 KB)  

    A new two-stage approach for brain image registration is proposed. In the first stage, an active contour algorithm is used to establish a length-preserving, one-to-one mapping between the cortical and the ventricular boundaries in the two images to be registered. This mapping is used in the second step by a two-dimensional transformation which is based on an elastic body deformation. This method was tested by registering magnetic resonance images to both photographic pathology images and atlas images View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Feature detection on 3-D images of dental imprints

    Publication Year: 1994 , Page(s): 287 - 296
    Cited by:  Papers (5)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (992 KB)  

    A computer vision approach for the extraction of feature points on 3D images of dental imprints is presented. The position of feature points are needed for the measurement of a set of parameters for automatic diagnosis of malocclusion problems in orthodontics. The system for the acquisition of the 3D profile of the imprint, the procedure for the detection of the interstices between teeth, and the approach for the identification of the type of tooth are described. The algorithm for the reconstruction of the surface of each type of tooth is presented. A new approach for the detection of feature points, called the watershed algorithm, is described in detail. The algorithm is a two-stage procedure which tracks the position of local minima at four different scales and produces a final map of the position of the minima. Experimental results of the application of the watershed algorithm on actual 3D images of dental imprints are presented for molars, premolars and canines. The segmentation approach for the analysis of the shape of incisors is also described in detail View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Study of computer diagnosis of X-ray and CT images in Japan-a brief survey

    Publication Year: 1994 , Page(s): 155 - 164
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1416 KB)  

    This short article was prepared to provide a very brief introduction to research activities in Japan in the field of computerised diagnosis using X-ray and CT images combined with picture processing and pattern understanding. After introducing 2 organizations of researchers in related fields, the author presents several examples of studies reported in domestic academic meetings and Japanese journals View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 3D displacement field reconstruction from planar tagged cardiac MR images

    Publication Year: 1994 , Page(s): 51 - 60
    Cited by:  Papers (6)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (620 KB)  

    A major problem in cardiac imaging is the measurement of cardiac motion for identification of ischemic and infarcted tissues. W.G. O'Dell et al. (J. Magn. Reson. Imag., vol.3(p), p.208, 1993) have recently proposed a method that uses magnetic resonance tag patterns to measure the 3D displacement field of the myocardium. The measurements are sparse, however, and interpolation is required to reconstruct a dense displacement field. Here, the authors propose a method for computing a dense displacement field on a regular 3D lattice from sparse displacement measurements. This method uses a multidimensional stochastic model for the true displacement field and the Fisher estimation framework to estimate a displacement vector at each point the lattice. Simulation results are presented that demonstrate the accuracy of the authors' technique View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A scalar function formulation for optical flow: applications to X-ray imaging

    Publication Year: 1994 , Page(s): 117 - 123
    Cited by:  Papers (7)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (564 KB)  

    The author presents results from a new formulation for determining velocities from a time-sequence of X-ray projection images of an incompressible fluid. Starting with the conservation of mass principle, and physics of X-ray projection, the author derives a motion constraint equation for projection imaging, a practical special case of which is shown to be the Horn and Schunck's optical flow constraint. The formulation is particularly efficient, as the flow field is obtained from a 90 degrees rotation applied to the gradient of a scalar function. It is shown that if specific criteria are met, in addition to normal flow which is commonly recoverable, the tangential component of optical flow is also recoverable, without the need for smoothness. An algorithm is presented to illustrate this. Preliminary results from the optical flow formulation applied to synthetic images, as well as contrast-injected X-ray images of flowing fluid, in a cylindrical phantom are presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Classification of nonrigid motion in 3D images using physics-based vibration analysis

    Publication Year: 1994 , Page(s): 61 - 69
    Cited by:  Papers (5)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (708 KB)  

    Presents a method for comparison and classification of nonrigid motion using a deformable model and the associated modal dynamics. Using the solid state physics formulation, the authors develop the equations of motion and the analytical expressions of vibration modes of a multidimensional elastically-deformable model that the authors use for tracking of nonrigid features in dynamic images. Thus, nonrigid motion of a 3D deformable object can be recovered in closed-form in modal space, by superimposing a few low-frequency modes that can be computed in a straightforward manner. The authors show that the deformation spectrum is a compact description of the deformation information, and allows deformation comparison and classification. Examples on 2D and 3D medical images are presented. The contribution of the paper is making modal analysis an efficient and easy-to-use tool for nonrigid motion classification View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The wrapper algorithm: surface extraction and simplification

    Publication Year: 1994 , Page(s): 204 - 213
    Cited by:  Papers (5)  |  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (868 KB)  

    The wrapper algorithm (Gueziec and Hummel, 1994) is a rapid and efficient algorithm for extracting and representing surfaces defined as isosurfaces in digitized volumetric data. Unlike the marching cubes algorithm (Lorensen and Cline, 1987), and in agreement with the Payne-Toga algorithm (1990), the authors further decompose the cubical mesh associated with the sampled data into tetrahedral cells. They guarantee the resulting surface representation to be closed and oriented, defined by a valid triangulation of the surface of the body, which in turn is presented as a collection of tetrahedra, some of which are only partly filled. The surface is extracted as a collection of closed triangles, where each triangle is an oriented closed curve contained within a single tetrahedron. The authors' method is an extension of the work of Doi and Koide (1991). In particular, the surface extraction process is followed by a simplification algorithm, which typically provides between 10 and 20 to 1 compression ratios in the surface representation while respecting a maximum distance bound to the original surface, and without noticeable degradation in surface fidelity. The authors display surfaces of the cranium from CT-scans and cortical surfaces from MR-scans at full resolution as well as synthetic surfaces, which have been extracted and simplified by their method View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multimodality medical imaging for radiotherapy treatment planning

    Publication Year: 1994 , Page(s): 235 - 244
    Cited by:  Papers (1)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (740 KB)  

    Radiation therapy is a treatment modality which seeks to deliver radiation energy to a localized site within a patient in order to destroy a malignant tumor. One of the most important aspects of radiation dose treatment planning is the accurate localization of tumor masses. An improved methodology is proposed for the delineation of tumors. Tumor localization is performed using a regularization approach which incorporates data fusion techniques. The contour information, derived from magnetic resonance and X-ray computed tomography, is used to enhance the delineation of tumor boundaries. Directional controlled continuity stabilizers are utilized to perform the fusion task View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multispectral analysis of the brain in magnetic resonance imaging

    Publication Year: 1994 , Page(s): 33 - 42
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (768 KB)  

    The authors demonstrate an improved differentiation of the most common tissue types in the human brain and surrounding structures by quantitative validation using multispectral analysis of magnetic resonance images. This is made possible by a combination of a special training technique and an increase in the number of magnetic resonance channel images with different pulse acquisition parameters. A description of the tissue-specific multivariate statistical distributions of the pixel intensity values and a statistical method to estimate the tissue-specific longitudinal and transverse relaxation times are given. It is concluded that multispectral analysis of magnetic resonance images is a valuable tool to recognize the most common normal tissue types in the brain and surrounding structures View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Deformable models for tagged MR images: reconstruction of two- and three-dimensional heart wall motion

    Publication Year: 1994 , Page(s): 317 - 323
    Cited by:  Papers (12)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (556 KB)  

    Magnetic resonance (MR) tissue tagging allows non-invasive in vivo measurement of soft tissue deformation. Planes of magnetic saturation are created, orthogonal to the imaging plane, which form dark lines (stripes) on the image. These stripes deform with the tissue and may be tracked in order to reconstruct the time-varying deformation of the structure. The authors have developed a semiautomatic line tracking algorithm to track the stripes to sub-pixel resolution throughout the deformation, based on active contour models (snakes). Each tracked stripe point gives displacement information orthogonal to the original tagging plane, i.e. a one dimensional constraint on the motion. Data from several tagging and imaging planes in different orientations was combined using a deformable model of the LV wall. Two or three dimensional motion and deformation was then reconstructed by fitting the data constraints by linear least squares View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Application of stereo techniques to angiography: qualitative and quantitative approaches

    Publication Year: 1994 , Page(s): 277 - 286
    Cited by:  Papers (1)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (672 KB)  

    The authors discuss the use of stereo imaging techniques as a low cost and low radiation dosage means to provide physicians with three-dimensional (3D) information for diagnosis and follow-up care of patients. Instead of using a pair of images obtained from perpendicular directions (as in the case of biplane radiography), the authors show that if a pair of small-angle stereo X-ray images are obtained, observers can vividly perceive the images in 3D during acquisition, i.e. real-time stereo visualization is possible. In addition, quantitative information such as blood flow measurements and the amount of arterial narrowing can be extracted off-line by using computational stereo algorithms. The potential advantages of using a small-angle stereo-based approach as opposed to biplane angiography are discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A distributed, object-oriented framework for medical image management and analysis: application to evaluation of medical image segmentation techniques

    Publication Year: 1994 , Page(s): 194 - 203
    Cited by:  Patents (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1124 KB)  

    A distributed framework for medical image management and analysis is described. Individual program modules interact over the network, creating and accessing a set of structural information resources that includes images, anatomic shape models, anatomic names, and contours which are the output of a knowledge-based segmentation program. Program modules are described that automatically evaluate the segmenter on large number of images, thereby allowing more robust testing of segmentation techniques than is currently possible in manual evaluations. It is suggested these kinds of programs will not only improve the evaluation of segmentation techniques, but will also enhance the overall utility of medical Picture Archiving and Communication Systems View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fitting of iso-surfaces using superquadrics and free-form deformations

    Publication Year: 1994 , Page(s): 184 - 193
    Cited by:  Papers (8)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (928 KB)  

    Recovery of 3D data with simple parametric models has been the subject of many studies over the last ten years. Many have used the notion of superquadrics introduced for graphics by Barr (1981). Different improvements were introduced to make the model a better representation of the data (Boult and Gross, 1987; Ferrie et al., 1989; Solina and Bajcsy 1990; Terzopoulos and Metaxas, 1991). The authors describe a two-steps method to fit a parametric deformable surface to 3D points. They suppose that a 3D image has been segmented to get a set of 3D points. The first step consists in their version of a superquadric fit with global tapering, similar to the method proposed by Boult and Gross (1987). The authors then make use of the technique of free-form deformations, as introduced by Sederberg and Parry (1986) in computer graphics. They present experimental results with synthetic and real 3D medical images View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Computer assisted diagnosis of lung cancer using helical X-ray CT

    Publication Year: 1994 , Page(s): 261 - 267
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (528 KB)  

    The authors describe a computer assisted diagnosis system for lung cancer that detects a candidate tumor in its early stage from helical X-ray CT images with precise measurement conditions. Recently, helical X-ray CT images have been used as a mass screening diagnostic tool, but considering the number of images to be checked, the checking time is too long. Here the authors describe a diagnosis system that can detect the candidate area for the lung cancer tumor in its early stage. The authors' diagnostic algorithm is based on image processing techniques and medical knowledge. From results of application to lung cancer patients, the authors describe the effectiveness of their algorithm View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Spatio-temporal tracking of material shape change via multi-dimensional splines

    Publication Year: 1994 , Page(s): 110 - 116
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (460 KB)  

    A new method for modeling shape changes in multidimensional image sequences of deforming objects is presented. The method consists of defining an orthogonal coordinate system on the segmented object in the first image frame and then calculating the deformations of this coordinate system over time. The deformations are found by minimizing an energy functional that consists of a linear combination of a data fidelity term and a shape-change constraint term. The shape-change constraint is based on the differential geometric properties of the parametrized material coordinate system. The deforming material coordinate system models an object's shape changes both locally and globally. The mathematics is first developed for an n-dimensional sequence of images. Examples are given for both 2D and 3D image sequences of both real and synthetic images View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Three-dimensional reconstruction of kidney from ultrasonic images

    Publication Year: 1994 , Page(s): 43 - 49
    Cited by:  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (448 KB)  

    A new system for three-dimensional kidney reconstruction is proposed. In this system, the authors have designed a rotational sono-probe to acquire renal sonographic images at four equally separated angles with respect to a fixed rotational axis. Then, on each acquired renal sonogram, a boundary detection algorithm based on elliptic deformable model is used to extract the kidney boundary. This model can consider the renal shape globally and incorporates many significant information, therefore, the kidney boundary can be segmented from poor quality sonogram successfully. The four extracted kidney borders are then used for reconstructing the 3D kidney by stacking up kidney contours estimated with spline curves. No complicated alignment is necessary in the authors' system. Experiments have shown good results in volume estimation for both in vivo and vitro studies View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Automatic extraction and measurement of leukocyte motion in microvessels using spatiotemporal image analysis

    Publication Year: 1994 , Page(s): 134 - 143
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (708 KB)  

    Describes a computer vision system for the automatic extraction of leukocytes that adhere to microvessel walls and accurate measurement of their velocities from a sequence of images. The motion of these leukocytes can be visualized as motion along the wall contours. The authors use the constraint that the leukocytes move along the vessel wall contours to generate a spatiotemporal image, and the leukocyte motion is then extracted using the methods of spatiotemporal image analysis. The generated spatiotemporal image is processed by a special-purpose orientation-selective filter and a subsequent grouping process newly developed for this application. The orientation-selective filter is designed by considering the particular properties of the spatiotemporal image in this application in order to enhance only the traces of leukocytes. In the subsequent grouping process, leukocyte trace segments are selected and grouped among all the segments obtained by simple thresholding and skeletonizing operations. The authors show experimentally that the proposed method can stably extract leukocyte motion, even when multiple leukocyte traces intersect View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A recursive filter for temporal analysis of cardiac motion

    Publication Year: 1994 , Page(s): 124 - 133
    Cited by:  Papers (8)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (648 KB)  

    A framework for temporal analysis of left ventricular (LV) endocardial wall motion is presented. This approach uses a novel technique of two-dimensional harmonic estimation to model the periodic nature of cardiac motion. A method for flow vector computation is presented which defines a relationship between image-derived, shape-based correspondences and a more desirable, smoothly varying, set of correspondences. A sequential filter is then constructed which takes into consideration this relationship as well as knowledge of temporal trends. Experimental results for contours derived from cycles of actual cardiac magnetic resonance images are presented. Applications to the analysis of regional LV wall function are discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Using optical flow to analyze the motion of human body organs from bioimages

    Publication Year: 1994 , Page(s): 100 - 109
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (748 KB)  

    It is evident that human body organs movement analysis requires the study of the dynamic evolution of an image sequence (2D or 3D), i.e. the analysis of a data flow versus time. In particular, the proposed approach is based on the computation of the “optical flow”, from two-dimensional and three-dimensional image sequences. This technique is applied to time sequences of magnetic resonance images to study knee joint biomechanics and heart wall motion. Several issues are discussed, related both to the computation of the optical flow field on 3D image sequences, and also concerning the relevance of the flow measurement for clinical diagnosis of pathologies. Moreover, the measurement of motion-related parameters, facilitates the planning for surgical intervention. Experiments performed on real data acquired from healthy patients and patients with heart and knee ligament pathologies are presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Matching 3-D anatomical surfaces with non-rigid deformations using octree-splines

    Publication Year: 1994 , Page(s): 144 - 153
    Cited by:  Papers (15)  |  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (880 KB)  

    Presents a new method for determining the minimal nonrigid deformation between two 3D surfaces, such as those which describe anatomical structures in 3D medical images. Although the authors match surfaces, they represent the deformation as a volumetric transformation. Their method performs a least squares minimization of the distance between the two surfaces of interest. To quickly and accurately compute distances between points on the two surfaces, the authors use a precomputed distance map represented using an octree spline whose resolution increases near the surface. To quickly and robustly compute the deformation, the authors use a second octree-spline to model the deformation function. The coarsest level of the deformation encodes the global (e.g., affine) transformation between the two surfaces, while finer levels encode smooth local displacements which bring the two surfaces into closer registration. The authors present experimental results on both synthetic and real 3D surfaces View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Computer vision and graphics in fluorescence microscopy

    Publication Year: 1994 , Page(s): 166 - 175
    Cited by:  Papers (1)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (968 KB)  

    The authors analyze medical images which are 3D (or 2D time series) images of individual cells, fluorescently labelled (sometimes dual-labelled) and imaged via a wide-field microscope. The goal is to visualize intracellular structures in these images, and to quantify the spatial relationships among these structures. A local, fast deformable model has been developed which finds the membrane of a cell. DAVE, the visualization tool, is a surface-based algorithm which converts voxel data into surfaces and hence permits the visualization of the cell membrane (as found by the deformable model) in conjunction with the image volume. It also permits the visualization of dual-labelled images and is optimized to emphasize the extent to which the labels colocalize. In addition, a statistical analysis of the significance of the colocalization is performed. The application of the developed tools to several biologically motivated problems is discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Intermodality 3D medical image registration with global search

    Publication Year: 1994 , Page(s): 225 - 234
    Cited by:  Papers (8)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (768 KB)  

    Describes a new registration method for three-dimensional medical images. It is important clinically to bring images from different modalities into alignment so that equivalent points can be identified. Often, functional images (showing metabolism or blood flow) are registered to structural images in order to more accurately interpret and quantify these images. This is especially important in areas of decreased function. This registration method determines the optimal transformation between the images using a match metric based on the distance transform of a structure visible in both modalities. The globally optimal transform is determined using a genetic optimization method and a hybrid technique using both genetic and gradient optimization. This provides a feasible way of determining the global solution making this method robust to local minima and insensitive to initial positioning View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On modal modeling for medical images: underconstrained shape description and data compression

    Publication Year: 1994 , Page(s): 70 - 79
    Cited by:  Papers (8)  |  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (808 KB)  

    The authors have previously described modal analysis, an efficient, physically-based solution for recovering, tracking, and recognizing solid models from 2D and 3D sensor data. The underlying representation consists of two levels: modal deformations, which describe the overall shape of a solid, and displacement maps, which employ a multiscale wavelet representation to provide local and fine surface detail. Here, the authors address the problem of recovering modal models in the underconstrained case of fitting a 3D model to contours found in medical slice and X-ray data. They describe an extension which can be used to incorporate measurement uncertainty while estimating the modal deformation parameters. Finally, the authors give details about how to compress dense 3D point data from surfaces, by use of displacement maps and wavelets View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new version of computer aided screening system in chest photofluorograms

    Publication Year: 1994 , Page(s): 307 - 316
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (664 KB)  

    Presents a new system for automated screening of chest photofluorograms taken with high voltage. This system detects and marks the shadows suspected to be abnormal using a procedure consisting of the following three steps: (1) detection of suspicious regions (SR) by a new filter (called Min-φDD filter), (2) measurement of features for each SR, and (3) classification of all SRs into two classes (normal or abnormal) by the maximum likelihood method. The Min-φDD filter enhances the abnormal shadows. The authors used three filters that have different sizes of masks considering the relation between the size of the filter and the size of the abnormal shadow. They calculated twenty-five features at each SR in Step (2) and designed three classifiers according to the size of the Min-φDD filter in Step (3). Fifty-eight chest photofluorograms taken with high voltage were used to evaluate the performance of the system. The recognition rates for the SR's evaluated by the L-method were from 67.5% to 75.5% using five features View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.