-
[show abstract]
[hide abstract]
ABSTRACT: In this paper, we propose a new classification method for five categories of lung tissues in high-resolution computed tomography (HRCT) images, with feature-based image patch approximation. We design two new feature descriptors for higher feature descriptiveness, namely the rotation-invariant Gaborlocal binary patterns (RGLBP) texture descriptor and multicoordinate histogram of oriented gradients (MCHOG) gradient descriptor. Together with intensity features, each image patch is then labeled based on its feature approximation from reference image patches. And a new patch-adaptive sparse approximation (PASA) method is designed with the following main components: minimum discrepancy criteria for sparse-based classification, patch-specific adaptation for discriminative approximation, and feature-space weighting for distance computation. The patchwise labelings are then accumulated as probabilistic estimations for region-level classification. The proposed method is evaluated on a publicly available ILD database, showing encouraging performance improvements over the state-of-the-arts.
IEEE transactions on medical imaging. 01/2013;
-
[show abstract]
[hide abstract]
ABSTRACT: We propose a novel joint probabilistic model that correlates a new probabilistic shape model with the corresponding global intensity distribution to segment multiple abdominal organs simultaneously. Our probabilistic shape model estimates the probability of an individual voxel belonging to the estimated shape of the object. The probability density of the estimated shape is derived from a combination of the shape variations of target class and the observed shape information. To better capture the shape variations, we used probabilistic principle component analysis optimized by expectation maximization to capture the shape variations and reduce computational complexity. The maximum a posteriori estimation was optimized by the iterated conditional mode-expectation maximization. We used 72 training datasets including low- and high-contrast CT images to construct the shape models for the liver, spleen and both kidneys. We evaluated our algorithm on 40 test datasets that were grouped into normal (34 normal cases) and pathologic (6 datasets) classes. The testing datasets were from different databases and manual segmentation was performed by different clinicians. We measured the volumetric overlap percentage error, relative volume difference, average square symmetric surface distance, false positive rate and false negative rate and our method achieved accurate and robust segmentation for multiple abdominal organs simultaneously.
IEEE transactions on information technology in biomedicine: a publication of the IEEE Engineering in Medicine and Biology Society 11/2012; · 1.69 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Analysis of primary lung tumors and disease in regional lymph nodes is important for lung cancer staging, and an automated system that can detect both types of abnormalities will be helpful for clinical routine. In this paper, we present a new method to automatically detect both tumors and abnormal lymph nodes simultaneously from positron emission tomography-computed tomography thoracic images. We perform the detection in a multistage approach, by first detecting all potential abnormalities, then differentiate between tumors and lymph nodes, and finally refine the detected tumors for false positive reduction. Each stage is designed with a discriminative model based on support vector machines and conditional random fields, exploiting intensity, spatial and contextual features. The method is designed to handle a wide and complex variety of abnormal patterns found in clinical datasets, consisting of different spatial contexts of tumors and abnormal lymph nodes. We evaluated the proposed method thoroughly on clinical datasets, and encouraging results were obtained.
IEEE transactions on medical imaging. 01/2012; 31(5):1061-75.
-
[show abstract]
[hide abstract]
ABSTRACT: Segmentation of multi-dimensional functional positron emission tomography (PET) studies into regions of interest (ROI) exhibiting similar temporal behavior is useful in diagnosis and evaluation of neurological images. Quantitative evaluation plays a crucial role in measuring the segmentation algorithm's performance. Due to the lack of "ground truth" available for evaluating segmentation of clinical images, automated segmentation results are usually compared with manual delineation of structures which is, however, subjective, and is difficult to perform. Alternatively, segmentation of co-registered anatomical images such as magnetic resonance imaging (MRI) can be used as the ground truth to the PET segmentation. However, this is limited to PET studies which have corresponding MRI. In this study, we introduce a framework for the objective and quantitative evaluation of functional PET study segmentation without the need for manual delineation or registration to anatomical images of the patient. The segmentation results are anatomically standardized to a functional brain atlas, where the segmentation of the corresponding MRI reference atlas image is used as the ground truth. We illustrate our evaluation framework by comparing the performance of two pixel-classification techniques based on k-means and fuzzy c-means cluster analysis, applied to clinical dynamic human brain PET studies. The experimental results show that the proposed evaluation framework is able to provide objective measures for segmentation comparison and performance.
Nuclear Science Symposium Conference Record, 2004 IEEE; 11/2004
-
[show abstract]
[hide abstract]
ABSTRACT: An unseeded, graph-theoretic segmentation algorithm based on Mumford-Shah energy minimization is applied to segmentation of brain FDG dynamic positron emission tomography data after preprocessing by principal component analysis, for the automated extraction of regions of interest, and, in particular, extraction of the internal carotid arteries and venous sinuses for the noninvasive estimation of the input plasma time activity curve. Evaluation on clinical FDG brain PET studies show that the internal carotids and venous sinuses can be robustly segmented in typical dynamic PET data sets, allowing for the fully automatic estimation of the arterial input curve.
Nuclear Science Symposium Conference Record, 2003 IEEE; 11/2003
-
[show abstract]
[hide abstract]
ABSTRACT: Optimal sampling schedule (OSS) is of great interest in biomedical experiment design, as it can improve the physiological parameter estimation precision and significantly reduce the samples required. A number of well designed algorithms and software packages have been developed, which deal with the instantaneous measurements at discrete times. However, in nuclear medicine tracer kinetic studies, the imaging systems, such as positron emission tomography (PET) and single photon emission computed tomography (SPECT), take measurements (images) based on continuous accumulation over time intervals. In this case, the existing algorithms cannot be used to design OSS so as to reduce the image frame numbers. In this paper, a general OSS design algorithm for the accumulative measurement is proposed. The potential usefulness of the algorithm is demonstrated by its designing OSS in [18F] fluoro-2-deoxy-D-glucose (FDG) studies with PET to estimate the local cerebral metabolic rate of glucose. The robustness of parameter estimation using the OSS with respect to intra-subject and inter-subject parameter variations is also presented.
Computer Methods and Programs in Biomedicine 05/2001; 65(1):45-59. · 1.52 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Dynamic imaging with positron emission tomography (PET) is widely used for the in vivo measurement of regional cerebral metabolic rate for glucose (rCMRGlc) with [18F]fluorodeoxy-D-glucose (FDG) and is used for the clinical evaluation of neurological disease. However, in addition to the acquisition of dynamic images, continuous arterial blood sampling is the conventional method to obtain the tracer time-activity curve in blood (or plasma) for the numeric estimation of rCMRGlc in mg glucose/100-g tissue/min. The insertion of arterial lines and the subsequent collection and processing of multiple blood samples are impractical for clinical PET studies because it is invasive, has the remote, but real potential for producing limb ischemia, and it exposes personnel to additional radiation and risks associated with handling blood. In this paper, based on our previously proposed method for extracting kinetic parameters from dynamic PET images, we developed a modified version (post-estimation method) to improve the numerical identifiability of the parameter estimates when we deal with data obtained from clinical studies. We applied both methods to dynamic neurologic FDG PET studies in three adults. We found that the input function and parameter estimates obtained with our noninvasive methods agreed well with those estimated from the gold standard method of arterial blood sampling and that rCMRGlc estimates were highly correlated (r = 0.973). More importantly, no significant difference was found between rCMRGlc estimated by our methods and the gold standard method (P > 0.16). We suggest that our proposed noninvasive methods may offer an advance over existing methods.
IEEE Transactions on Information Technology in Biomedicine 03/2001; 5(1):67-76. · 1.68 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Functional imaging with dynamic positron emission tomography (PET) has been playing a crucial and expanding role in biomedical research and clinical diagnosis, providing image-wide quantitative and qualitative physiological functions in the human body, and supporting visualization of the distribution of these functions corresponding to anatomical structures. A number of parametric imaging algorithms have been developed. We give a brief study on some existing and our recently, developed techniques for generating parametric images. An integrated system for functional image data processing and visualization, and a Web-based application are presented
Computer Graphics International, 2000. Proceedings; 02/2000
-
[show abstract]
[hide abstract]
ABSTRACT: The original generalized linear least squares (GLLS) algorithm was developed for non-uniformly sampled biomedical system parameter estimation using finely sampled instantaneous measurements (D. Feng, S.C. Huang, Z. Wang, D. Ho, An unbiased parametric imaging algorithm for non-uniformly sampled biomedical system parameter estimation, IEEE Trans. Med. Imag. 15 (1996) 512-518). This algorithm is particularly useful for image-wide generation of parametric images with positron emission tomography (PET), as it is computationally efficient and statistically reliable (D. Feng, D. Ho, Chen, K., L.C. Wu, J.K. Wang, R.S. Liu, S.H. Yeh, An evaluation of the algorithms for determining local cerebral metabolic rates of glucose using positron emission tomography dynamic data, IEEE Trans. Med. Imag. 14 (1995) 697-710). However, when dynamic PET image data are sampled according to the optimal image sampling schedule (OISS) to reduce memory and storage space (X. Li, D. Feng, K. Chen, Optimal image sampling schedule: A new effective way to reduce dynamic image storage space and functional image processing time, IEEE Trans. Med. Imag. 15 (1996) 710-718), only a few temporal image frames are recorded (e.g. only four images are recorded for the four parameter fluoro-deoxy-glucose (FDG) model). These image frames are recorded in terms of accumulated radio-activity counts and as a result, the direct application of GLLS is not reliable as instantaneous measurement samples can no longer be approximated by averaging of accumulated measurements over the sampling intervals. In this paper, we extend GLLS to OISS-GLLS which deals with the fewer accumulated measurement samples obtained from OISS dynamic systems. The theory and algorithm of this new technique are formulated and studied extensively. To investigate statistical reliability and computational efficiency of OISS-GLLS, a simulation study using dynamic PET data was performed. OISS-GLLS using 4-measurement samples was compared to the non-linear least squares (NLS) method using 22-measurement samples, GLLS using 22-measurement samples and OISS-NLS using 4-measurement samples. Results demonstrated that OISS-GLLS was able to achieve parameter estimates of equivalent accuracy and reliability in comparison to NLS or GLLS using finely sampled measurements (22-measurement samples), or OISS-NLS using optimally sampled measurements (4-measurement samples). Further more, as fewer measurement samples are used in OISS-GLLS, this algorithm is computationally faster than NLS or GLLS. Therefore, OISS-GLLS is well-suited for image-wide parameter estimation when PET image data are recorded according to the optimal image sampling schedule.
Computer Methods and Programs in Biomedicine 05/1999; 59(1):31-43. · 1.52 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Two rapid estimation algorithms for construction of cerebral blood flow (CBF) and oxygen utilization (CMRO) images with dynamic positron emission tomography (PET) are presented. These algorithms are based on the linear least squares (LLS) and generalized linear least squares (GLLS) methodologies. Using the conventional two-compartmental model and multiple tracer studies, we derived a linear relationship for brain tissue activity to arterial blood activity, time-integrated arterial blood activity and time-integrated brain tissue activity. The LLS technique is computationally efficient as no regression analysis is required, while GLLS is used to refine the estimates obtained from LLS. A comparative study using non-linear least squares regression (NLS) revealed excellent correlation between the new algorithms for various noise levels expected in clinical applications. A sensitivity analysis was performed to examine reliability and identifiability of the parameter estimates. In view of the results, LLS and GLLS provide rapid and reliable estimates of CBF and CMRO when applied to dynamic PET data. These algorithms are particularly suitable for pixel-by-pixel construction of high resolution and highly accurate PET functional images.
Computer Methods and Programs in Biomedicine 03/1999; 58(2):99-117. · 1.52 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: When performing dynamic studies using emission tomography the tracer distribution changes during acquisition of a single set of projections. This is particularly true for some positron emission tomography (PET) systems which, like single photon emission computed tomography (SPECT), acquire data over a limited angle at any time, with full projections obtained by rotation of the detectors. In this paper, an approach is proposed for processing data from these systems, applicable to either PET or SPECT. A method of interpolation, based on overlapped parabolas, is used to obtain an estimate of the total counts in each pixel of the projections for each required frame-interval, which is the total time to acquire a single complete set of projections necessary for reconstruction. The resultant projections are reconstructed using traditional filtered backprojection (FBP) and tracer kinetic parameters are estimated using a method which relies on counts integrated over the frame-interval rather than instantaneous values. Simulated data were used to illustrate the technique's capabilities with noise levels typical of those encountered in either PET or SPECT. Dynamic datasets were constructed, based on kinetic parameters for fluoro-deoxy-glucose (FDG) and use of either a full ring detector or rotating detector acquisition. For the rotating detector, use of the interpolation scheme provided reconstructed dynamic images with reduced artefacts compared to unprocessed data or use of linear interpolation. Estimates for the metabolic rate of glucose had similar bias to those obtained from a full ring detector.
IEEE Transactions on Medical Imaging 01/1999; 17(6):986-94. · 3.64 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: The recently developed generalized linear least squares (GLLS) algorithm has been found very useful in non-uniformly sampled biomedical signal processing and parameter estimation. However, the current version of the algorithm cannot deal with signals and systems containing repeated eigenvalues. In this paper, we extend the algorithm, so that it can be used for non-uniformly sampled signals and systems with/without repeated eigenvalues. The related theory and detailed derivation of the algorithm are given. A case study is presented, which demonstrates that the extended algorithm can provide more choices for system identification and is able to select the most suitable model for the system from the non-uniformly sampled noisy signal.
Computer Methods and Programs in Biomedicine 12/1998; 57(3):167-77. · 1.52 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: The authors developed and tested a method for the noninvasive quantification of the cerebral metabolic rate for glucose (CMRglc) using positron emission tomography (PET), 18F-fluoro-2-deoxyglucose, the Patlak method, and an image-derived input function. Dynamic PET data acquired 12 to 48 seconds after rapid tracer injection were summed to identify carotid artery regions of interest (ROIs). The input function then was generated from the carotid artery ROIs. To correct spillover, the early summed image was superimposed over the last PET frame, a tissue ROI was drawn around the carotid arteries, and a tissue time activity curve (TAC) was generated. Three venous samples were drawn from the tracer injection site at a later time and used for the spillover and partial volume correction by non-negative least squares method. Twenty-six patient data sets were studied. It was found that the image-derived input function was comparable in shape and magnitude to the one obtained by arterial blood sampling. Moreover, no significant difference was found between CMRglc estimated by the Patlak method using either the arterial blood sampling data or the image-derived input function.
Journal of Cerebral Blood Flow & Metabolism 08/1998; 18(7):716-23. · 5.01 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: With the recent development in scatter and attenuation correction algorithms, dynamic single photon emission computerized tomography (SPECT) can potentially yield physiological parameters, with tracers exhibiting suitable kinetics such as thallium-201 (Tl-201). A systematic way is proposed to investigate the minimum data acquisition times and sampling requirements for estimating physiological parameters with quantitative dynamic SPECT. Two different sampling schemes were investigated with Monte Carlo simulations: 1) Continuous data collection for total study duration ranging from 30-240 min. 2) Continuous data collection for first 10-45 min followed by a delayed study at approximately 3 h. Tissue time activity curves with realistic noise were generated from a mean plasma time activity curve and rate constants (K1 - k4) derived from Tl-201 kinetic studies in 16 dogs. Full dynamic sampling schedules (DynSS) were compared to optimum sampling schedules (OSS). We found that OSS can reliably estimate the blood flow related K1 and Vd comparable to DynSS. A 30-min continuous collection was sufficient if only K1 was of interest. A split session schedule of a 30-min dynamic followed by a static study at 3 h allowed reliable estimation of both K1 and Vd avoiding the need for a prolonged (>60-min) continuous dynamic acquisition. The methodology developed should also be applicable to optimizing sampling schedules for other SPECT tracers.
IEEE Transactions on Medical Imaging 06/1998; 17(3):334-43. · 3.64 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Positron emission tomography (PET) provides the ability to extract useful quantitative information not available through other radiological techniques. In certain studies, the physiological parameters of interest cannot be determined from the data obtained from a single PET experiment alone. In this case, multiple experiments are required. At present, the methods used to analyse measurements acquired from multiple experiments often involve considering them separately during the modelling procedures. These methods of analysis may cause errors to be propagated through successive modelling procedures and do not fully utilise the information content provided by the PET measurements. A new method is presented, based on linear least squares for the analysis of PET dynamic data acquired from multiple experiments. This method simultaneously considers the complete set of measurements obtained and provides reliable parameter estimates. The efficient use of the information content provided by multiple experiments is considered and the propagation of errors is discussed. To facilitate our discussion, we apply this new method to the estimation of the cerebral metabolic rate of oxygen and the parameters of the oxygen utilisation model as a practical example. The results demonstrate a significant improvement in the reliability and estimation accuracy of the estimates for this new method. Furthermore, this method reduced the likelihood of errors being propagated. Therefore, the proposed method is suitable for the analysis of multiple PET dynamic datasets.
Medical & Biological Engineering & Computing 02/1998; 36(1):83-90. · 1.88 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: The validation study is described of a new modelling method that has been developed, using tracer kinetic modelling with positron emission tomography (PET) to achieve non-invasive measurement of myocardial metabolic rate of glucose (MMRGlc). Eight data sets obtained from dynamic cardiac PET 2-[18F]fluoro-2-deoxy-D-glucose (FDG) studies on human subjects are employed, and the estimation of MMRGlc using both the new and traditional methods is compared. The results from all eight human FDG studies are consistent with those from previous computer simulations. With the new method, the estimated mean of K (a parameter directly proportional to MMRGlc) increases by about 8%, and that of k 4 (the rate constant of FDG dephosphorylation) decreases by about 48%. The approach should be more suitable for use in dynamic cardiac PET studies when non-invasive means are used to obtain the plasma time-activity curve from left-ventricle PET images.
Medical & Biological Engineering & Computing 02/1998; 36(1):112-7. · 1.88 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Positron emission tomography (PET) is an important tool for enabling quantification of human brain function. However, quantitative studies using tracer kinetic modeling require the measurement of the tracer time-activity curve in plasma (PTAC) as the model input function. It is widely believed that the insertion of arterial lines and the subsequent collection and processing of the biomedical signal sampled from the arterial blood are not compatible with the practice of clinical PET, as it is invasive and exposes personnel to the risks associated with the handling of patient blood and radiation dose. Therefore, it is of interest to develop practical noninvasive measurement techniques for tracer kinetic modeling with PET. In this paper, a technique is proposed to extract the input function together with the physiological parameters from the brain dynamic images alone. The identifiability of this method is tested rigorously by using Monte Carlo simulation. The results show that the proposed method is able to quantify all the required parameters by using the information obtained from two or more regions of interest (ROI's) with very different dynamics in the PET dynamic images. There is no significant improvement in parameter estimation for the local cerebral metabolic rate of glucose (LCMRGlc) if the number of ROI's are more than three. The proposed method can provide very reliable estimation of LCMRGlc, which is our primary interest in this study.
IEEE Transactions on Information Technology in Biomedicine 01/1998; 1(4):243-54. · 1.68 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: An algorithm for diagnostically lossless dynamic image data
compression is proposed. The theory and implementation of this algorithm
are presented. Taking advantage of domain specific knowledge related to
medical imaging and medical practice, we achieve very high compression
ratios. An example using the fluoro-deoxy-glucose tracer and dynamic
positron emission tomography is presented to evaluate the performance of
the proposed algorithm. As a result of our study, the storage space for
dynamic image data can be reduced by more than 95%, without loss in
diagnostic quality
Intelligent Processing Systems, 1997. ICIPS '97. 1997 IEEE International Conference on; 11/1997
-
[show abstract]
[hide abstract]
ABSTRACT: In this study, we propose a method and investigate the reduction of dynamic image data with positron emission tomography (PET). The method is based upon the use of sampling schedules with a reduced number of scanning intervals and the use of an integral model in the cost function of nonlinear regression. The application of this method is illustrated by the problem of estimating the metabolic rate of glucose with the [18F]2-fluoro-2-deoxyglucose (FDG) model. Computer simulations were performed using various sampling schedules with scanning intervals of different lengths. The results were compared in terms of the accuracy and precision of the estimated parameters. It has been found that the use of sampling schedules with a reduced number of scanning intervals in conjunction with the integral model is very effective. The number of images in dynamic PET FDG studies can be reduced by a factor of 4.5 without losing the accuracy and precision of the parameter estimates.
Computer Methods and Programs in Biomedicine 07/1997; 53(2):71-80. · 1.52 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Whole-body positron emission tomography (PET) has recently emerged as an important imaging tool for cancer detection and staging. Initial applications of the technique have been primarily qualitative. One of the major reasons is the limits imposed by kinetically undersampled data over the whole body, as opposed to the standard method of continuous dynamic sampling in one body location. In this paper, a new estimation method using weighted nonlinear least squares (WNLS) for the first bed position and Bayesian regression (BR) for subsequent positions is proposed. A general criterion for designing optimal sampling schedules which maximizes the measurement information with multiple bed positions is developed. The overall approach is illustrated with the problem of estimating the metabolic rate of glucose (MRGLu) in tumors at different axial positions (image bed positions) in the body by using computer simulations and patient data. The results show that estimates of MRGLu using sparse data and the optimized Bayesian approach are comparable with those obtained by standard methods and fully sampled data. This study demonstrates the potential of the technique described for quantification where several bed positions have to be used to image all the regions of interest (ROI).
IEEE Transactions on Biomedical Engineering 11/1996; 43(10):1021-8. · 2.28 Impact Factor
