[Show abstract][Hide abstract] ABSTRACT: This paper addresses the automated segmentation of multiple organs in upper abdominal computed tomography (CT) data. The aim of our study is to develop methods to effectively construct the conditional priors and use their prediction power for more accurate segmentation as well as easy adaptation to various imaging conditions in CT images, as observed in clinical practice. We propose a general framework of multi-organ segmentation which effectively incorporates interrelations among multiple organs and easily adapts to various imaging conditions without the need for supervised intensity information. The features of the framework are as follows: (1) A method for modeling conditional shape and location (shape-location) priors, which we call prediction-based priors, is developed to derive accurate priors specific to each subject, which enables the estimation of intensity priors without the need for supervised intensity information. (2) Organ correlation graph is introduced, which defines how the conditional priors are constructed and segmentation processes of multiple organs are executed. In our framework, predictor organs, whose segmentation is sufficiently accurate by using conventional single-organ segmentation methods, are pre-segmented, and the remaining organs are hierarchically segmented using conditional shape-location priors. The proposed framework was evaluated through the segmentation of eight abdominal organs (liver, spleen, left and right kidneys, pancreas, gallbladder, aorta, and inferior vena cava) from 134 CT data from 86 patients obtained under six imaging conditions at two hospitals. The experimental results show the effectiveness of the proposed prediction-based priors and the applicability to various imaging conditions without the need for supervised intensity information. Average Dice coefficients for the liver, spleen, and kidneys were more than 92%, and were around 73% and 67% for the pancreas and gallbladder, respectively.
Medical image analysis 07/2015; 26(1). DOI:10.1016/j.media.2015.06.009 · 3.65 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: In the following we will present a newly developed X-ray calibration phantom and its integration for 2D/3D pelvis reconstruction and subsequent automatic cup planning. Two different planning strategies were applied and evaluated with clinical data.
Two different cup planning methods were investigated: The first planning strategy is based on a combined pelvis and cup statistical atlas. Thereby, the pelvis part of the combined atlas is matched to the reconstructed pelvis model, resulting in an optimized cup planning. The second planning strategy analyzes the morphology of the reconstructed pelvis model to determine the best fitting cup implant.
The first planning strategy was compared to 3D CT-based planning. Digitally reconstructed radiographs of THA patients with differently severe pathologies were used to evaluate the accuracy of predicting the cup size and position. Within a discrepancy of one cup size, the size was correctly identified in 100% of the cases for Crowe type I datasets and in 77.8% of the cases for Crowe type II, III and IV datasets. The second planning strategy was analyzed with respect to the eventually implanted cup size. In seven patients, the estimated cup diameter was correct within one cup size, while the estimation for the remaining five patients differed by two cup sizes.
While both planning strategies showed the same prediction rate with a discrepancy of one cup size (87.5%), the prediction of the exact cup size was increased for the statistical atlas-based strategy (56%) in contrast to the anatomically driven approach (37.5%).
The proposed approach demonstrated the clinical validity of using 2D/3D reconstruction technique for cup planning.
[Show abstract][Hide abstract] ABSTRACT: Purpose:
Determination of acetabular cartilage loss in the hip joint is a clinically significant metric that requires image segmentation. A new semiautomatic method to segment acetabular cartilage in computed tomography (CT) arthrography scans was developed and tested.
A semiautomatic segmentation method was developed based on the combination of anatomical and statistical information. Anatomical information is identified using the pelvic bone position and the contact area between cartilage and bone. Statistical information is acquired from CT intensity modeling of acetabular cartilage and adjacent tissue structures. This method was applied to the identification of acetabular cartilages in 37 intra-articular CT arthrography scans.
The semiautomatic anatomical-statistical method performed better than other segmentation methods. The semiautomatic method was effective in noisy scans and was able to detect damaged cartilage.
The new semiautomatic method segments acetabular cartilage by fully utilizing the statistical and anatomical information in CT arthrography datasets. This method for hip joint cartilage segmentation has potential for use in many clinical applications.
International Journal of Computer Assisted Radiology and Surgery 07/2014; 10(4). DOI:10.1007/s11548-014-1101-1 · 1.71 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Purpose:
Total knee arthroplasty (TKA) 3D kinematic analysis requires 2D/3D image registration of X-ray fluoroscopic images and a computer-aided design (CAD) model of the knee implant. However, these techniques cannot provide information on the radiolucent polyethylene insert, since the insert silhouette does not appear clearly in X-ray images. Therefore, it is difficult to obtain the 3D kinematics of the polyethylene insert, particularly the mobile-bearing insert. A technique for 3D kinematic analysis of a mobile-bearing insert used in TKA was developed using X-ray fluoroscopy. The method was tested and a clinical application was evaluated.
Tantalum beads and a CAD model of the mobile-bearing TKA insert are used for 3D pose estimation of the mobile-bearing insert used in TKA using X-ray fluoroscopy. The insert model was created using four identical tantalum beads precisely located at known positions in a polyethylene insert using a specially designed insertion device. Finally, the 3D pose of the insert model was estimated using a feature-based 2D/3D registration technique, using the silhouette of beads in fluoroscopic images and the corresponding CAD insert model. In vitro testing for the repeatability of the positioning of the tantalum beads and computer simulations for 3D pose estimation of the mobile-bearing insert were performed.
The pose estimation accuracy achieved was sufficient for analyzing mobile-bearing TKA kinematics (RMS error: within 1.0 mm and 1.0°, except for medial-lateral translation). In a clinical application, nine patients with mobile-bearing TKA were investigated and analyzed with respect to a deep knee bending motion.
A 3D kinematic analysis technique was developed that enables accurate quantitative evaluation of mobile-bearing TKA kinematics. This method may be useful for improving implant design and optimizing TKA surgical techniques.
International Journal of Computer Assisted Radiology and Surgery 06/2014; 10(4). DOI:10.1007/s11548-014-1093-x · 1.71 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Liver-shape analysis and quantification is still an open research subject. Quantitative assessment of the liver is of clinical importance in various procedures such as diagnosis, treatment planning, and monitoring. Liver-shape classification is of clinical importance for corresponding intra-subject and inter-subject studies. In this research, we propose a novel technique for the liver-shape classification based on Spherical Harmonics (SH) coefficients. The proposed liver-shape classification algorithm consists of the following steps: (a) Preprocessing, including mesh generation and simplification, point-set matching, and surface to template alignment; (b) Liver-shape parameterization, including surface normalization, SH expansion followed by parameter space registration; (c) Feature selection and classification, including frequency based feature selection, feature space reduction by Principal Component Analysis (PCA), and classification. The above multi-step approach is novel in the sense that registration and feature selection for liver-shape classification is proposed and implemented and validated for the normal and diseases liver in the SH domain. Various groups of SH features after applying conventional PCA and/or ordered by p-value PCA are employed in two classifiers including Support Vector Machine (SVM) and k-Nearest Neighbor (k-NN) in the presence of 101 liver data sets. Results show that the proposed specific features combined with classifiers outperform existing liver-shape classification techniques that employ liver surface information in the spatial domain. In the available data sets, the proposed method can successful classify normal and diseased livers with a correct classification rate of above 90 %. The performed result in average is higher than conventional liver-shape classification method. Several standard metrics such as Leave-one-out cross-validation and Receiver Operating Characteristic (ROC) analysis are employed in the experiments and confirm the effectiveness of the proposed liver-shape classification with respect to conventional techniques.
Journal of Medical Systems 05/2014; 38(5):20. DOI:10.1007/s10916-014-0020-6 · 2.21 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Statistical shape models (SSMs) represent morphological variations of a specific object. When there are large shape variations, the shape parameters constitute a large space that may include incorrect parameters. The human liver is a non-rigid organ subject to large deformations due to external forces or body position changes during scanning procedures. We developed and tested a population-based model to represent the shape of liver.
Upper abdominal CT-scan input images are represented by a conventional shape model. The shape parameters of individual livers extracted from the CT scans are employed to classify them into different populations. Corresponding to each population, an SSM model is built. The liver surface parameter space is divided into several subspaces which are more compact than the original space. The proposed model was tested using 29 CT-scan liver image data sets. The method was evaluated by model compactness, reconstruction error, generality and specificity measures.
The proposed model is implemented and tested using CT scans that included liver shapes with large shape variations. The method was compared with conventional and recently developed shape modeling methods. The accuracy of the proposed model was nearly twice that achieved with the conventional model. The proposed population-based model was more general compared with the conventional model. The mean reconstruction error of the proposed model was 0.029 mm while that of the conventional model was 0.052 mm.
A population-based model to represent the shape of liver was developed and tested with favorable results. Using this approach, the liver shapes from CT scans were modeled by a more compact, more general, and more accurate model.
International Journal of Computer Assisted Radiology and Surgery 04/2014; DOI:10.1007/s11548-014-1000-5 · 1.71 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Purpose:
We quantified prostate swelling and the intraprostatic point shift during high intensity focused ultrasound using real-time ultrasound.
Materials and methods:
The institutional review board approved this retrospective study. Whole gland high intensity focused ultrasound was done in 44 patients with clinically localized prostate cancer. Three high intensity focused ultrasound sessions were required to cover the entire prostate, including the anterior zone (session 1), middle zone (session 2) and posterior zone (session 3). Computer assisted 3-dimensional reconstructions based on 3 mm step-section images of intraoperative transrectal ultrasound were compared before and after each session.
Most prostate swelling and intraprostatic point shifts occurred during session 1. The median percent volume increase was 18% for the transition zone, 9% for the peripheral zone and 13% for the entire prostate. The volume percent increase in the transition zone (p <0.001), peripheral zone (p = 0.001) and entire prostate (p = 0.001) statistically depended on the volume of each area measured preoperatively. The median 3-dimensional intraprostatic shift was 3.7 mm (range 0.9 to 13) in the transition zone and 5.5 mm (range 0.2 to 14) in the peripheral zone. A significant negative linear correlation was found between the preoperative presumed circle area ratio, and the percent increase in prostate volume (p = 0.001) and shift (p = 0.01) during high intensity focused ultrasound.
We quantified significant prostate swelling and shift during high intensity focused ultrasound. Smaller prostates and a smaller preoperative presumed circle area ratio were associated with greater prostate swelling and intraprostatic shifts. Real-time intraoperative adjustment of the treatment plan impacts the achievement of precise targeting during high intensity focused ultrasound, especially in prostates with a smaller volume and/or a smaller preoperative presumed circle area ratio.
The Journal of urology 04/2013; 190(4). DOI:10.1016/j.juro.2013.03.116 · 4.47 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: In this paper, we present an algorithm to segment the liver in low-contrast CT images. As the first step of our algorithm, we define a search range for the liver boundary. Then, the EM algorithm is utilized to estimate parameters of a 'Gaussian Mixture' model that conforms to the intensity distribution of the liver. Using the statistical parameters of the intensity distribution, we introduce a new thresholding technique to classify image pixels. We assign a distance feature vectors to each pixel and segment the liver by a K-means clustering scheme. This initial boundary of the liver is conditioned by the Fourier transform. Then, a Geodesic Active Contour algorithm uses the boundaries to find the final surface. The novelty in our method is the proper selection and combination of sub-algorithms so as to find the border of an object in a low-contrast image. The number of parameters in the proposed method is low and the parameters have a low range of variations. We applied our method to 30 datasets including normal and abnormal cases of low-contrast/high-contrast images and it was extensively evaluated both quantitatively and qualitatively. Minimum of Dice similarity measures of the results is 0.89. Assessment of the results proves the potential of the proposed method for segmentation in low-contrast images.
IEICE Transactions on Information and Systems 04/2013; E96.D(4):798-807. DOI:10.1587/transinf.E96.D.798 · 0.21 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: This paper reviews methods for computer assisted medical intervention using statistical models and machine learning technologies, which would be particularly useful for representing prior information of anatomical shape, motion, and deformation to extrapolate intraoperative sparse data as well as surgeons' expertise and pathology to optimize interventions. Firstly, we present a review of methods for recovery of static anatomical structures by only using intraoperative data without any preoperative patient-specific information. Then, methods for recovery of intraoperative motion and deformation are reviewed by combining intraoperative sparse data with preoperative patient-specific stationary data, which is followed by a survey of articles which incorporated biomechanics. Furthermore, the articles are reviewed which addressed the used of statistical models for optimization of interventions. Finally, we conclude the survey by describing the future perspective.
IEICE Transactions on Information and Systems 04/2013; E96.D(4):784-797. DOI:10.1587/transinf.E96.D.784 · 0.21 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The paper addresses the automated segmentation of multiple organs in upper abdominal CT data. We propose a framework of multi-organ segmentation which is adaptable to any imaging conditions without using intensity information in manually traced training data. The features of the framework are as follows: (1) the organ correlation graph (OCG) is introduced, which encodes the spatial correlations among organs inherent in human anatomy; (2) the patient-specific organ shape and location priors obtained using OCG enable the estimation of intensity priors from only target data and optionally a number of untraced CT data of the same imaging condition as the target data. The proposed methods were evaluated through segmentation of eight abdominal organs (liver, spleen, left and right kidney, pancreas, gallbladder, aorta, and inferior vena cava) from 86 CT data obtained by four imaging conditions at two hospitals. The performance was comparable to the state-of-the-art method using intensity priors constructed from manually traced data.
[Show abstract][Hide abstract] ABSTRACT: We describe a method to capture disease-specific components in organ shapes. A statistical shape model, constructed by the principal component analysis (PCA) of organ shapes, is used to define the subspace representing inter-subject shape variability. The first PCA is applied to the datasets of healthy organ shapes to define the subspace of normal variability. Then, the datasets of diseased shapes are projected onto the orthogonal complement (OC) of the sub-space of normal variability, and the second PCA is applied to the projected datasets to derive the subspace representing the disease-specific variability. To calculate the OC of an n-dimensional subspace, a novel closed-form formulation is developed. Experiments were performed to show that the support vector machine classification in the OC subspace better discriminated healthy and diseased liver shapes using 99 CT data. The effects of the number of training data and the difference in segmentation methods on the classification accuracy were evaluated to clarify the characteristics of the proposed method.
[Show abstract][Hide abstract] ABSTRACT: Segmentation of the femur and pelvis is a prerequisite for patient-specific planning and simulation for hip surgery. Accurate boundary determination of the femoral head and acetabulum is the primary challenge in diseased hip joints because of deformed shapes and extreme narrowness of the joint space. To overcome this difficulty, we investigated a multi-stage method in which the hierarchical hip statistical shape model (SSM) is initially utilized to complete segmentation of the pelvis and distal femur, and then the conditional femoral head SSM is used under the condition that the regions segmented during the previous stage are known. CT data from 100 diseased patients categorized on the basis of their disease type and severity, which included 200 hemi-hips, were used to validate the method, which delivered significantly increased segmentation accuracy for the femoral head.
[Show abstract][Hide abstract] ABSTRACT: Automated segmentation of multiple organs in CT data of the upper abdomen is addressed. In order to explicitly incorporate the spatial interrelations among organs, we propose a method for finding and representing the interrelations based on canonical correlation analysis. Furthermore, methods are developed for constructing and utilizing the statistical atlas in which inter-organ constraints are explicitly incorporated to improve accuracy of multi-organ segmentation. The proposed methods were tested to perform segmentation of eight abdominal organs (liver, spleen, kidneys, pancreas, gallbladder, aorta, and inferior vena cava) from various imaging conditions of CT datasets. 87 datasets acquired at two institutions were used for the validation. Significant accuracy improvement was observed for several organs in comparison with the conventional method.
Conference proceedings: ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference 08/2012; 2012:3986-9. DOI:10.1109/EMBC.2012.6346840
[Show abstract][Hide abstract] ABSTRACT: Atlas-based methods for automated preoperative planning of the femoral stem implant in total hip arthroplasty are described. Statistical atlases are constructed from a number of past preoperative plans prepared by experienced surgeons in order to represent the surgeon's expertise of the planning. Two types of atlases are considered. One is a statistical distance map atlas, which represents surgeon's preference of the contact pattern between the femoral canal (host bone) and stem (implant) surfaces. The other is an optimal reference plan, which is selected as the best representative plan expected to minimize the deviation from the surgeon's preferred contact pattern. These atlases are fitted to the patient data to automatically generate the preoperative plan of the femoral stem. In this paper, we formulate a general framework of atlas-based implant planning, and then describe the methods for construction and utilization of the two proposed atlases. In the experiments, we used 40 cases to evaluate the proposed methods and compare them with previous methods by defining the errors as differences between automated and surgeon's plans. By using the proposed methods, the positional and orientation errors were significantly reduced compared with the previous methods and the size error was superior to inter-surgeon variability in size selection using 2D templates on an X-ray image reported in previous work.
Medical image analysis 11/2011; 16(2):415-26. DOI:10.1016/j.media.2011.10.005 · 3.65 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The automated segmentation of multiple organs in CT data of the upper abdomen is addressed. In order to explicitly incorporate the spatial interrelations among organs, we propose a method for finding and representing the interrelations based on canonical correlation analysis. Furthermore, methods are developed for constructing and utilizing the statistical atlas in which inter-organ constraints are explicitly incorporated to improve accuracy of multi-organ segmentation. The proposed methods were tested to perform segmentation of seven abdominal organs (liver, spleen, kidneys, pancreas, gallbladder and inferior vena cava) from contrast-enhanced CT datasets and was compared to a previous approach. 28 datasets acquired at two institutions were used for the validation. Significant accuracy improvement was observed for the segmentation of pancreas and gallbladder while there was no accuracy reduction for any organ.
Proceedings of the Third international conference on Abdominal Imaging: computational and Clinical Applications; 09/2011