Segmentation of fluid-filled structures, such as the urinary bladder, from three-dimensional ultrasound images is necessary for measuring their volume. This paper describes a system for image enhancement, segmentation and volume measurement of fluid-filled structures on 3D ultrasound images. The system was applied for the measurement of urinary bladder volume. Results show an average error of less than 10% in the estimation of the total bladder volume.
Volumetric analysis of the brain from MR images is an important biomedical research tool. Segmentation of the brain parenchyma and its constituent tissue types, the gray matter and the white matter, is necessary for volumetric information in longitudinal and cross-sectional studies. We have implemented and compared two different classes of algorithms for segmentation of the brain parenchyma. In the first algorithm a combination of automatic thresholding and 3-D mathematical morphology was used to segment the brain while in the second algorithm an optical flow-based 3-D non-rigid registration approach was used to warp an MR head atlas to the subject brain. For tissue classification within the brain area a 3-D Markov Random Field model was used in conjunction with supervised and unsupervised classification. The algorithms described above were validated on a data set provided at the Internet Brain Segmentation Repository that consists of 20 normal T1 volumes (3 mm slice thickness) with manually segmented brain and manually classified tissues. While the morphological segmentation algorithm had an average similarity index of 0.918, the atlas-based brain segmentation algorithm has an average similarity index of 0.953. The supervised tissue classification had an average similarity index of 0.833 for gray matter voxels and 0.766 for white matter voxels. The performance of these algorithms is quite acceptable to end-users both in terms of accuracy and speed.
Many medical imaging modalities produce spatial or temporal stacks of image data. Segmentation of such image stacks has many applications ranging from quantitative measurements to surgical and radiation treatment planning. The key idea presented in this paper is that of propagating information serially from one slice to the next within an interactive framework. Since information on adjacent slices is very similar, segmentation on one slice can be propagated with slight modification to adjacent slices. The segmentation algorithms that we have developed within this framework are all based on energy minimization principles with an additional constraint that the segmentation on a given image slice is similar to the segmentation predicted from the previous image slice. An optical flow approach is used to predict segmentation from one slice to the next. Three types of algorithms have been developed within the above paradigm for different applications --(1) A Mumford and Shah energy- minimizing algorithm combining edge and region information in a region-growing framework, (2) an active contour model-based tracking method, and (3) an algorithm based on pixel classification and Markov random fields. We recognize the fact that interactivity is very important in medical image segmentation. Therefore, our segmentation tools are available in a Java-based graphical user interface (GUI), allowing users to initialize various segmentation algorithms or to edit the results of automatic segmentation, if desired.
Implantation of radioactive isotopes within the prostate for the treatment of early stage localized prostate cancer is becoming a popular treatment option. Postoperative calculation of the dose delivered to the prostate requires accurate verification of the number and location of seeds within the prostate. Current post operative dosimetry technique requires the dosimetrist to manually count and record the position of each seed from x-ray computed tomography (CT) images. This procedure is operator-dependent and time-consuming, thus limiting the ability of different brachytherapy centers to compare results and create a standard methodology. Seed identification is performed by thresholding the CT images interactively, using a graphical user interface, followed by mathematical morphology to remove noise. Segmented seeds are grouped into regions via connected-component analysis. Regions are then classified into seeds using a prior knowledge of the seed dimensions and their relative positions in the consecutive CT images. Unresolved regions, which can indicate the presence of more than one seed, are corrected manually. The efficiency of this tool was evaluated by comparing the time to manually count the seeds to the time required to do the same task using the automated program. For 15 sets of images from 15 patients, the average time for manually counting the seeds was 45 minutes per patient versus 6.4 minutes on average per patients, the average time for manually counting the seeds was 45 minutes per patient versus 6.4 minutes on average per patient when the software was used to perform the same task. Using the interactive visualization and segmentation algorithm, the time required to count the seeds during post- implant dosimetry has been reduced by a factor of 7 compared to the existing manual technique.
Prostate brachytherapy is a treatment procedure for localized prostate cancer. It involves placing needles and subsequently radioactive seeds under ultrasound guidance into predetermined targets within the prostate. As the pubic arch can be a barrier to successful placement of the needles, preoperative assessment requires visualization of the pubic arch with respect to the prostate. Current CT- based techniques to assess pubic arch interference (PAI) are expensive and time-consuming. This paper describes a new technique using transrectal ultrasound that enables the visualization of the pubic arch bone and the prostate gland simultaneously. The technique involves speckle suppression in the pubic arch ultrasound image and contrast enhancement of the pubic bones using sticks algorithm. This step is followed by noise filtering using percentile thresholding and curve fitting. The detected arch is superimposed on the transverse cross-sectional image of the prostate at its largest position predicted by the algorithm was compared with the 'true' pubic arch position determined at surgery by placing needles into multiple coordinates corresponding and adjacent to the predicted arch position. The accuracy of the algorithm in detecting the pubic arch was tested on 50 patients. Of 1030 points tested, the algorithm prediction was correct at 932 points. The mean Type II error, i.e., the algorithm predicted soft tissue while bone was encountered during needle insertion, was 2.9 percent, which corresponds to less than 1 out of 22 test points along the predicted pubic arch. The accuracy of our algorithm is good and the errors are within clinically-acceptable limits.
In this paper, we study the problem of estimating and segmenting the optical flow field in image sequences. A variational framework based on the Mumford-Shah functional is introduced for simultaneous edge preserved optical flow estimation and motion-based segmentation. The proposed energy functional for optical flow field and its corresponding edge set if formulated to have three additive terms. The first and second terms measure the deviation from the optical flow constraints over the whole image and its smoothness at all the non-edge locations in L2 norm, respectively, while the third term regularizes the total length of all the edges. The minimization of this functional is carried out by the vector graduated nonconvexity (VGNC) algorithm with the gradient descent iterating scheme. This framework is then extended to fuse spatio-temporal segmentation by adding two more terms for spatial segmentation in the above formulation. One term is the L2 difference between the original image and an approximation of the image, while the other is the regularization of approximate image at all non-edge locations. The same VGNC procedure is performed to minimized the functional to obtain the optical flow field, the piecewise smooth image, and the spatio-temporal edge image. We illustrate the presented method and its numerical implementation on tactical image sequences.
With the number of men seeking medical care for prostate diseases rising steadily, the need of a fast and accurate prostate boundary detection and volume estimation tool is being increasingly experienced by the clinicians. Currently, these measurements are made manually, which results in a large examination time. A possible solution is to improve the efficiency by automating the boundary detection and volume estimation process with minimal involvement from the human experts. In this paper, we present an algorithm based on SNAKES to detect the boundaries. Our approach is to selectively enhance the contrast along the edges using an algorithm called sticks and integrate it with a SNAKES model. This integrated algorithm requires an initial curve for each ultrasound image to initiate the boundary detection process. We have used different schemes to generate the curves with a varying degree of automation and evaluated its effects on the algorithm performance. After the boundaries are identified, the prostate volume is calculated using planimetric volumetry. We have tested our algorithm on 6 different prostate volumes and compared the performance against the volumes manually measured by 3 experts. With the increase in the user inputs, the algorithm performance improved as expected. The results demonstrate that given an initial contour reasonably close to the prostate boundaries, the algorithm successfully delineates the prostate boundaries in an image, and the resulting volume measurements are in close agreement with those made by the human experts.
Medical image segmentation has many applications, including tumor localization, radiation therapy planting, and 3D modeling, but its current use falls far short of its potential. To address this shortcoming, we are developing a unified software environment that facilitates the development and deployment of new and existing medical image segmentation algorithms, including classification-based, shape-based, region-based, edge-based, and hybrid algorithms.
Duplicate documents are frequently found in large databases of digital documents, such as those found in digital libraries or in the government declassification effort. Efficient duplicate document detection is important not only to allow querying for similar documents, but also to filter out redundant information in large document databases. We have designed three different algorithm to identify duplicate documents. The first algorithm is based on features extracted from the textual content of a document, the second algorithm is based on wavelet features extracted from the document image itself, and the third algorithm is a combination of the first two. These algorithms are integrated within the DocBrowse system for information retrieval from document images which is currently under development at MathSoft. DocBrowse supports duplicate document detection by allowing (1) automatic filtering to hide duplicate documents, and (2) ad hoc querying for similar or duplicate documents. We have tested the duplicate document detection algorithms on 171 documents and found that text-based method has an average 11-point precision of 97.7 percent while the image-based method has an average 11- point precision of 98.9 percent. However, in general, the text-based method performs better when the document contains enough high-quality machine printed text while the image- based method performs better when the document contains little or no quality machine readable text.
The goal of this research is to implement an extensible, user-friendly, portable and affordable medical image analysis environment that frees the analyst from having to deal with low-level programming issues. With this environment, the user can rapidly prototype, test, validate and statistically analyze new medical image analysis algorithms. The environment tightly integrates a medical image database that supports image content and metadata based querying; visualization and browsing of query results; S-Plus, a powerful object-oriented, interpreted programming environment, with a large suite of built-in visualization, data analysis, statistics and image processing tools. The system interfaces with any SQL-based relational database management system. The database schema is based on the DICOM standard. Using MedPlus, the user can develop image analysis procedures that integrate database queries, interactive image analysis, statistical inference and database archival of results using ether a scripting language.
KEYWORDS: Ultrasonography, Image processing, Head, Digital signal processing, Detection and tracking algorithms, Signal processing, Multimedia, Algorithm development, Sun, Fetus
Ultrasound as a medical imaging modality offers the clinician a real-time of the anatomy of the internal organs/tissues, their movement, and flow noninvasively. One of the applications of ultrasound is to monitor fetal growth by measuring biparietal diameter (BPD) and head circumference (HC). We have been working on automatic detection of fetal head boundaries in ultrasound images. These detected boundaries are used to measure BPD and HC. The boundary detection algorithm is based on active contour models and takes 32 seconds on an external high-end workstation, SUN SparcStation 20/71. Our goal has been to make this tool available within an ultrasound machine and at the same time significantly improve its performance utilizing multimedia technology. With the advent of high- performance programmable digital signal processors (DSP), the software solution within an ultrasound machine instead of the traditional hardwired approach or requiring an external computer is now possible. We have integrated our boundary detection algorithm into a programmable ultrasound image processor (PUIP) that fits into a commercial ultrasound machine. The PUIP provides both the high computing power and flexibility needed to support computationally-intensive image processing algorithms within an ultrasound machine. According to our data analysis, BPD/HC measurements made on PUIP lie within the interobserver variability. Hence, the errors in the automated BPD/HC measurements using the algorithm are on the same order as the average interobserver differences. On PUIP, it takes 360 ms to measure the values of BPD/HC on one head image. When processing multiple head images in sequence, it takes 185 ms per image, thus enabling 5.4 BPD/HC measurements per second. Reduction in the overall execution time from 32 seconds to a fraction of a second and making this multimedia system available within an ultrasound machine will help this image processing algorithm and other computer-intensive imaging applications become a practical tool for the sonographers in the feature.
KEYWORDS: Expectation maximization algorithms, Monte Carlo methods, Statistical analysis, Data modeling, Statistical modeling, Image segmentation, Medical imaging, Mathematical modeling, Stochastic processes, Data analysis
Deformable models have gained much popularity recently for many applications in medical imaging, such as image segmentation, image reconstruction, and image registration. Such models are very powerful because various kinds of information can be integrated together in an elegant statistical framework. Each such piece of information is typically associated with a user-defined parameter. The values of these parameters can have a significant effect on the results generated using these models. Despite the popularity of deformable models for various applications, not much attention has been paid to the estimation of these parameters. In this paper we describe systematic methods for the automatic estimation of these deformable model parameters. These methods are derived by posing the deformable models as a Bayesian inference problem. Our parameter estimation methods use Markov chain Monte Carlo methods for generating samples from highly complex probability distributions.
In this paper, we propose a methodology for evaluating medical image segmentation algorithms where the only information available is boundaries outlined by multiple expert observers. In this case, the results of the segmentation algorithm can be evaluated against the multiple observers' outlines. We have derived statistics to enable us to find whether the computer-generated boundaries agree with the observers' hand-outlined boundaries as much as the different observers agree with each other. We illustrate the use of this methodology by evaluating image segmentation algorithms on two different applications in ultrasound imaging. In the first application, we attempt to find the epicardial and endocardial boundaries from cardiac ultrasound images, and in the second application, our goal is to find the fetal skull boundaries from prenatal ultrasound images.
We propose a segmentation approach which integrates region growing and edge detection in a regularization framework. Our method is a modified active contour model and uses region statistics in addition to gradient information. We formulate the active contour model using a Bayesian approach. We have implemented this integrated approach and characterized its performance on synthetic images and on 36 short-axis cardiac ultrasound images. The resulting boundaries are compared to true boundaries in the case of the synthetic images and to manually outlined boundaries in the case of ultrasound images. The results are also compared with those obtained using the balloon force to expand the active contour model. We found that our integrated algorithm detects boundaries more accurately than the active contour method using a balloon force. Furthermore, the integrated algorithm is less sensitive to the placement of the initial contour inside the LV cavity than the active contour algorithm using a balloon force.
Accurate identification of boundaries of the left ventricle lets the cardiologist determine important physiological parameters like the left-ventricular ejection fraction, volume of the left ventricle, and regional heart wall thickening; all of which aid in better diagnosis of heart diseases. We have developed a new semi-automated method to determine the left-ventricular boundaries from short-axis echocardiograms. Our method is based on the active contour models, also know as snakes, originally proposed by Kass et al. Our method was tested on images obtained from 18 patients; manual outlining was used as reference for comparison. Our results were also compared to Detmer et at., who used the same images to test their algorithm. The errors in detecting boundaries by using our algorithm were found to be within the reproducibilty of manual outlining. We also implemented a 3D extension of the active contour algorithm, where the third dimension is time, and are currently working on a clinical validation of this algorithm. The 3D algorithm partly alleviates the problems encountered in the 2D algorithm due to missing boundaries in echocardiograms.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.