An adaptive-focus statistical shape model for segmentation and shape modeling of 3-D brain structures

Implementation of needle-tracking technology for real-time transrectal ultrasound-guided interstitial gynecological HDR brachytherapy: A feasibility study

PURPOSE

To investigate the feasibility of adapting a commercial prostate biopsy system for transrectal ultrasound (TRUS)-guided hybrid gynecological (GYN) high-dose-rate (HDR) brachytherapy (BT). Leveraging 3D-TRUS and MR image fusion, the prototype system aims to improve real-time needle placement accuracy.

MATERIALS AND METHODS

A second-generation, multi-imaging modality female pelvic phantom was developed to validate the system's feasibility. Software and hardware modifications, including user-accessible calibration modules and a redesigned needle guide for multi-needle insertion, were made to the pre-existing commercial system to enable use for GYN BT applications. An end-to-end feasibility test was performed to acquire 3D-TRUS images, perform contour-based registration with pre-implant MR, and insert six needles to targeted locations under real-time TRUS guidance. A 30° tandem without ovoids was added to mimic a hybrid GYN implant. The most proximal and most distal distances between the planned needle track and the visible portion of each inserted needle were measured. A CT/MR image-based treatment plan with a prescribed dose of 6 Gy was generated for the resulting 3D-TRUS-guided implant (tandem and needles) within the phantom.

RESULTS

The modified phantom improved needle visualization and insertion range by de-gassing the silicone and increasing the window size. The system accuracy for average ± standard deviations from intended needle tracks was 1.31 ± 1.36 mm (proximal) and 2.04 ± 2.05 mm (distal). Post-implant imaging confirmed needle placement and good target coverage. Needle placement was verified on CT/MR images and treatment plan quality was clinically acceptable.

CONCLUSIONS

With enhanced needle placement accuracy and integrated clinical workflow, this study demonstrates the feasibility of adapting a commercially available prostate biopsy system for 3D-TRUS-guided hybrid GYN HDR BT.

Learning Canonical Embeddings for Unsupervised Shape Correspondence With Locally Linear Transformations

We present a new approach to unsupervised shape correspondence learning between pairs of point clouds. We make the first attempt to adapt the classical locally linear embedding algorithm (LLE)-originally designed for nonlinear dimensionality reduction-for shape correspondence. The key idea is to find dense correspondences between shapes by first obtaining high-dimensional neighborhood-preserving embeddings of low-dimensional point clouds and subsequently aligning the source and target embeddings using locally linear transformations. We demonstrate that learning the embedding using a new LLE-inspired point cloud reconstruction objective results in accurate shape correspondences. More specifically, the approach comprises an end-to-end learnable framework of extracting high-dimensional neighborhood-preserving embeddings, estimating locally linear transformations in the embedding space, and reconstructing shapes via divergence measure-based alignment of probability density functions built over reconstructed and target shapes. Our approach enforces embeddings of shapes in correspondence to lie in the same universal/canonical embedding space, which eventually helps regularize the learning process and leads to a simple nearest neighbors approach between shape embeddings for finding reliable correspondences. Comprehensive experiments show that the new method makes noticeable improvements over state-of-the-art approaches on standard shape correspondence benchmark datasets covering both human and nonhuman shapes.

A novel registration-based algorithm for prostate segmentation via the combination of SSM and CNN

PURPOSE

Precise determination of target is an essential procedure in prostate interventions, such as prostate biopsy, lesion detection, and targeted therapy. However, the prostate delineation may be tough in some cases due to tissue ambiguity or lack of partial anatomical boundary. In this study, we proposed a novel supervised registration-based algorithm for precise prostate segmentation, which combine the convolutional neural network (CNN) with a statistical shape model (SSM).

METHODS

The proposed network mainly consists of two branches. One called SSM-Net branch was exploited to predict the shape transform matrix, shape control parameters, and shape fine-tuning vector, for the generation of the prostate boundary. Furtherly, according to the inferred boundary, a normalized distance map was calculated as the output of SSM-Net. Another branch named ResU-Net was employed to predict a probability label map from the input images at the same time. Integrating the output of these two branches, the optimal weighted sum of the distance map and the probability map was regarded as the prostate segmentation.

RESULTS

Two public datasets PROMISE12 and NCI-ISBI 2013 were utilized to evaluate the performance of the proposed algorithm. The results demonstrate that the segmentation algorithm achieved the best performance with an SSM of 9500 nodes, which obtained a dice of 0.907 and an average surface distance of 1.85 mm. Compared with other methods, our algorithm delineates the prostate region more accurately and efficiently. In addition, we verified the impact of model elasticity augmentation and the fine-tuning item on the network segmentation capability. As a result, both factors have improved the delineation accuracy, with dice increased by 10% and 7% respectively.

CONCLUSIONS

Our segmentation method has the potential to be an effective and robust approach for prostate segmentation. This article is protected by copyright. All rights reserved.

Investigating the Effects of Demographics on Shoulder Morphology and Density Using Statistical Shape and Density Modeling

A better understanding of how the shape and density of the shoulder vary among members of a population can help design more effective population-based orthopedic implants. The main objective of this study was to develop statistical shape models (SSMs) and statistical density models (SDMs) of the shoulder to describe the main modes of variability in the shape and density distributions of shoulder bones within a population in terms of principal components (PCs). These PC scores were analyzed, and significant correlations were observed between the shape and density distributions of the shoulder and demographics of the population, such as sex and age. Our results demonstrated that when the overall body sizes of male and female donors were matched, males still had, on average, larger scapulae and thicker humeral cortical bones. Moreover, we concluded that age has a weak but significant inverse effect on the density within the entire shoulder. Weak and moderate, but significant, correlations were also found between many modes of shape and density variations in the shoulder. Our results suggested that donors with bigger humeri have bigger scapulae and higher bone density of humeri corresponds with higher bone density in the scapulae. Finally, asymmetry, to some extent, was noted in the shape and density distributions of the contralateral bones of the shoulder. These results can be used to help guide the designs of population-based prosthesis components and pre-operative surgical planning.

Association of multiparametric MRI quantitative imaging features with prostate cancer gene expression in MRI-targeted prostate biopsies

Standard clinicopathological variables are inadequate for optimal management of prostate cancer patients. While genomic classifiers have improved patient risk classification, the multifocality and heterogeneity of prostate cancer can confound pre-treatment assessment. The objective was to investigate the association of multiparametric (mp)MRI quantitative features with prostate cancer risk gene expression profiles in mpMRI-guided biopsies tissues.Global gene expression profiles were generated from 17 mpMRI-directed diagnostic prostate biopsies using an Affimetrix platform. Spatially distinct imaging areas ('habitats') were identified on MRI/3D-Ultrasound fusion. Radiomic features were extracted from biopsy regions and normal appearing tissues. We correlated 49 radiomic features with three clinically available gene signatures associated with adverse outcome. The signatures contain genes that are over-expressed in aggressive prostate cancers and genes that are under-expressed in aggressive prostate cancers. There were significant correlations between these genes and quantitative imaging features, indicating the presence of prostate cancer prognostic signal in the radiomic features. Strong associations were also found between the radiomic features and significantly expressed genes. Gene ontology analysis identified specific radiomic features associated with immune/inflammatory response, metabolism, cell and biological adhesion. To our knowledge, this is the first study to correlate radiogenomic parameters with prostate cancer in men with MRI-guided biopsy.

Skeletal Shape Correspondence Through Entropy

We present a novel approach for improving the shape statistics of medical image objects by generating correspondence of skeletal points. Each object's interior is modeled by an s-rep, i.e., by a sampled, folded, two-sided skeletal sheet with spoke vectors proceeding from the skeletal sheet to the boundary. The skeleton is divided into three parts: the up side, the down side, and the fold curve. The spokes on each part are treated separately and, using spoke interpolation, are shifted along that skeleton in each training sample so as to tighten the probability distribution on those spokes' geometric properties while sampling the object interior regularly. As with the surface/boundary-based correspondence method of Cates et al., entropy is used to measure both the probability distribution tightness and the sampling regularity, here of the spokes' geometric properties. Evaluation on synthetic and real world lateral ventricle and hippocampus data sets demonstrate improvement in the performance of statistics using the resulting probability distributions. This improvement is greater than that achieved by an entropy-based correspondence method on the boundary points.

Automatic construction of statistical shape models using deformable simplex meshes with vector field convolution energy

BACKGROUND

In the active shape model framework, principal component analysis (PCA) based statistical shape models (SSMs) are widely employed to incorporate high-level a priori shape knowledge of the structure to be segmented to achieve robustness. A crucial component of building SSMs is to establish shape correspondence between all training shapes, which is a very challenging task, especially in three dimensions.

METHODS

We propose a novel mesh-to-volume registration based shape correspondence establishment method to improve the accuracy and reduce the computational cost. Specifically, we present a greedy algorithm based deformable simplex mesh that uses vector field convolution as the external energy. Furthermore, we develop an automatic shape initialization method by using a Gaussian mixture model based registration algorithm, to derive an initial shape that has high overlap with the object of interest, such that the deformable models can then evolve more locally. We apply the proposed deformable surface model to the application of femur statistical shape model construction to illustrate its accuracy and efficiency.

RESULTS

Extensive experiments on ten femur CT scans show that the quality of the constructed femur shape models via the proposed method is much better than that of the classical spherical harmonics (SPHARM) method. Moreover, the proposed method achieves much higher computational efficiency than the SPHARM method.

CONCLUSIONS

The experimental results suggest that our method can be employed for effective statistical shape model construction.

Learning Non-rigid Deformations for Robust, Constrained Point-based Registration in Image-Guided MR-TRUS Prostate Intervention

Accurate and robust non-rigid registration of pre-procedure magnetic resonance (MR) imaging to intra-procedure trans-rectal ultrasound (TRUS) is critical for image-guided biopsies of prostate cancer. Prostate cancer is one of the most prevalent forms of cancer and the second leading cause of cancer-related death in men in the United States. TRUS-guided biopsy is the current clinical standard for prostate cancer diagnosis and assessment. State-of-the-art, clinical MR-TRUS image fusion relies upon semi-automated segmentations of the prostate in both the MR and the TRUS images to perform non-rigid surface-based registration of the gland. Segmentation of the prostate in TRUS imaging is itself a challenging task and prone to high variability. These segmentation errors can lead to poor registration and subsequently poor localization of biopsy targets, which may result in false-negative cancer detection. In this paper, we present a non-rigid surface registration approach to MR-TRUS fusion based on a statistical deformation model (SDM) of intra-procedural deformations derived from clinical training data. Synthetic validation experiments quantifying registration volume of interest overlaps of the PI-RADS parcellation standard and tests using clinical landmark data demonstrate that our use of an SDM for registration, with median target registration error of 2.98 mm, is significantly more accurate than the current clinical method. Furthermore, we show that the low-dimensional SDM registration results are robust to segmentation errors that are not uncommon in clinical TRUS data.

Abdomen and spinal cord segmentation with augmented active shape models

Active shape models (ASMs) have been widely used for extracting human anatomies in medical images given their capability for shape regularization of topology preservation. However, sensitivity to model initialization and local correspondence search often undermines their performances, especially around highly variable contexts in computed-tomography (CT) and magnetic resonance (MR) images. In this study, we propose an augmented ASM (AASM) by integrating the multiatlas label fusion (MALF) and level set (LS) techniques into the traditional ASM framework. Using AASM, landmark updates are optimized globally via a region-based LS evolution applied on the probability map generated from MALF. This augmentation effectively extends the searching range of correspondent landmarks while reducing sensitivity to the image contexts and improves the segmentation robustness. We propose the AASM framework as a two-dimensional segmentation technique targeting structures with one axis of regularity. We apply AASM approach to abdomen CT and spinal cord (SC) MR segmentation challenges. On 20 CT scans, the AASM segmentation of the whole abdominal wall enables the subcutaneous/visceral fat measurement, with high correlation to the measurement derived from manual segmentation. On 28 3T MR scans, AASM yields better performances than other state-of-the-art approaches in segmenting white/gray matter in SC.

Constrained Statistical Modelling of Knee Flexion From Multi-Pose Magnetic Resonance Imaging

Reconstruction of the anterior cruciate ligament (ACL) through arthroscopy is one of the most common procedures in orthopaedics. It requires accurate alignment and drilling of the tibial and femoral tunnels through which the ligament graft is attached. Although commercial computer-assisted navigation systems exist to guide the placement of these tunnels, most of them are limited to a fixed pose without due consideration of dynamic factors involved in different knee flexion angles. This paper presents a new model for intraoperative guidance of arthroscopic ACL reconstruction with reduced error particularly in the ligament attachment area. The method uses 3D preoperative data at different flexion angles to build a subject-specific statistical model of knee pose. To circumvent the problem of limited training samples and ensure physically meaningful pose instantiation, homogeneous transformations between different poses and local-deformation finite element modelling are used to enlarge the training set. Subsequently, an anatomical geodesic flexion analysis is performed to extract the subject-specific flexion characteristics. The advantages of the method were also tested by detailed comparison to standard Principal Component Analysis (PCA), nonlinear PCA without training set enlargement, and other state-of-the-art articulated joint modelling methods. The method yielded sub-millimetre accuracy, demonstrating its potential clinical value.

Locally-constrained boundary regression for segmentation of prostate and rectum in the planning CT images

Automatic and accurate segmentation of the prostate and rectum in planning CT images is a challenging task due to low image contrast, unpredictable organ (relative) position, and uncertain existence of bowel gas across different patients. Recently, regression forest was adopted for organ deformable segmentation on 2D medical images by training one landmark detector for each point on the shape model. However, it seems impractical for regression forest to guide 3D deformable segmentation as a landmark detector, due to large number of vertices in the 3D shape model as well as the difficulty in building accurate 3D vertex correspondence for each landmark detector. In this paper, we propose a novel boundary detection method by exploiting the power of regression forest for prostate and rectum segmentation. The contributions of this paper are as follows: (1) we introduce regression forest as a local boundary regressor to vote the entire boundary of a target organ, which avoids training a large number of landmark detectors and building an accurate 3D vertex correspondence for each landmark detector; (2) an auto-context model is integrated with regression forest to improve the accuracy of the boundary regression; (3) we further combine a deformable segmentation method with the proposed local boundary regressor for the final organ segmentation by integrating organ shape priors. Our method is evaluated on a planning CT image dataset with 70 images from 70 different patients. The experimental results show that our proposed boundary regression method outperforms the conventional boundary classification method in guiding the deformable model for prostate and rectum segmentations. Compared with other state-of-the-art methods, our method also shows a competitive performance.

Methodology to study the three-dimensional spatial distribution of prostate cancer and their dependence on clinical parameters

A methodology to study the relationship between clinical variables [e.g., prostate specific antigen (PSA) or Gleason score] and cancer spatial distribution is described. Three-dimensional (3-D) models of 216 glands are reconstructed from digital images of whole mount histopathological slices. The models are deformed into one prostate model selected as an atlas using a combination of rigid, affine, and B-spline deformable registration techniques. Spatial cancer distribution is assessed by counting the number of tumor occurrences among all glands in a given position of the 3-D registered atlas. Finally, a difference between proportions is used to compare different spatial distributions. As a proof of concept, we compare spatial distributions from patients with PSA greater and less than [Formula: see text] and from patients older and younger than 60 years. Results suggest that prostate cancer has a significant difference in the right zone of the prostate between populations with PSA greater and less than [Formula: see text]. Age does not have any impact in the spatial distribution of the disease. The proposed methodology can help to comprehend prostate cancer by understanding its spatial distribution and how it changes according to clinical parameters. Finally, this methodology can be easily adapted to other organs and pathologies.

Three-dimensional nonrigid landmark-based magnetic resonance to transrectal ultrasound registration for image-guided prostate biopsy

Registration of three-dimensional (3-D) magnetic resonance (MR) to 3-D transrectal ultrasound (TRUS) prostate images is an important step in the planning and guidance of 3-D TRUS guided prostate biopsy. In order to accurately and efficiently perform the registration, a nonrigid landmark-based registration method is required to account for the different deformations of the prostate when using these two modalities. We describe a nonrigid landmark-based method for registration of 3-D TRUS to MR prostate images. The landmark-based registration method first makes use of an initial rigid registration of 3-D MR to 3-D TRUS images using six manually placed approximately corresponding landmarks in each image. Following manual initialization, the two prostate surfaces are segmented from 3-D MR and TRUS images and then nonrigidly registered using the following steps: (1) rotationally reslicing corresponding segmented prostate surfaces from both 3-D MR and TRUS images around a specified axis, (2) an approach to find point correspondences on the surfaces of the segmented surfaces, and (3) deformation of the surface of the prostate in the MR image to match the surface of the prostate in the 3-D TRUS image and the interior using a thin-plate spline algorithm. The registration accuracy was evaluated using 17 patient prostate MR and 3-D TRUS images by measuring the target registration error (TRE). Experimental results showed that the proposed method yielded an overall mean TRE of [Formula: see text] for the rigid registration and [Formula: see text] for the nonrigid registration, which is favorably comparable to a clinical requirement for an error of less than 2.5 mm. A landmark-based nonrigid 3-D MR-TRUS registration approach is proposed, which takes into account the correspondences on the prostate surface, inside the prostate, as well as the centroid of the prostate. Experimental results indicate that the proposed method yields clinically sufficient accuracy.

Three-Dimensional Nonrigid MR-TRUS Registration Using Dual Optimization

In this study, we proposed an efficient nonrigid magnetic resonance (MR) to transrectal ultrasound (TRUS) deformable registration method in order to improve the accuracy of targeting suspicious regions during a three dimensional (3-D) TRUS guided prostate biopsy. The proposed deformable registration approach employs the multi-channel modality independent neighborhood descriptor (MIND) as the local similarity feature across the two modalities of MR and TRUS, and a novel and efficient duality-based convex optimization-based algorithmic scheme was introduced to extract the deformations and align the two MIND descriptors. The registration accuracy was evaluated using 20 patient images by calculating the TRE using manually identified corresponding intrinsic fiducials in the whole gland and peripheral zone. Additional performance metrics [Dice similarity coefficient (DSC), mean absolute surface distance (MAD), and maximum absolute surface distance (MAXD)] were also calculated by comparing the MR and TRUS manually segmented prostate surfaces in the registered images. Experimental results showed that the proposed method yielded an overall median TRE of 1.76 mm. The results obtained in terms of DSC showed an average of 80.8±7.8% for the apex of the prostate, 92.0±3.4% for the mid-gland, 81.7±6.4% for the base and 85.7±4.7% for the whole gland. The surface distance calculations showed an overall average of 1.84±0.52 mm for MAD and 6.90±2.07 mm for MAXD.

Quantitative imaging: quantification of liver shape on CT using the statistical shape model to evaluate hepatic fibrosis

RATIONALE AND OBJECTIVES

To investigate the usefulness of the statistical shape model (SSM) for the quantification of liver shape to evaluate hepatic fibrosis.

MATERIALS AND METHODS

Ninety-one subjects (45 men and 46 women; age range, 20-75 years) were included in this retrospective study: 54 potential liver donors and 37 patients with chronic liver disease. The subjects were classified histopathologically according to the fibrosis stage as follows: F0 (n = 55); F1 (n = 6); F2 (3); F3 (n = 1); and F4 (n = 26). Each subject underwent contrast-enhanced computed tomography (CT) using a 64-channel scanner (0.625-mm slice thickness). An abdominal radiologist manually traced the liver boundaries on every CT section using an image workstation; the boundaries were used for subsequent analyses. An SSM was constructed by the principal component analysis of the subject data set, which defined a parametric model of the liver shapes. The shape parameters were calculated by fitting SSM to the segmented liver shape of each subject and were used for the training of a linear support vector regression (SVR), which classifies the liver fibrosis stage to maximize the area under the receiver operating characteristic curve (AUC). SSM/SVR models were constructed and were validated in a leave-one-out manner. The performance of our technique was compared to those of two previously reported types of caudate-right lobe ratios (C/RL-m and C/RL-r).

RESULTS

In our SSM/SVR models, the AUC values for the classification of liver fibrosis were 0.96 (F0 vs. F1-4), 0.95 (F0-1 vs. F2-4), 0.96 (F0-2 vs. F3-4), and 0.95 (F0-3 vs. F4). These values were significantly superior to AUC values using the C/RL-m or C/RL-r ratios (P < .005).

CONCLUSIONS

SSM was useful for estimating the stage of hepatic fibrosis by quantifying liver shape.

Review of ultrasound image guidance in external beam radiotherapy: I. Treatment planning and inter-fraction motion management

In modern radiotherapy, verification of the treatment to ensure the target receives the prescribed dose and normal tissues are optimally spared has become essential. Several forms of image guidance are available for this purpose. The most commonly used forms of image guidance are based on kilovolt or megavolt x-ray imaging. Image guidance can also be performed with non-harmful ultrasound (US) waves. This increasingly used technique has the potential to offer both anatomical and functional information.This review presents an overview of the historical and current use of two-dimensional and three-dimensional US imaging for treatment verification in radiotherapy. The US technology and the implementation in the radiotherapy workflow are described. The use of US guidance in the treatment planning process is discussed. The role of US technology in inter-fraction motion monitoring and management is explained, and clinical studies of applications in areas such as the pelvis, abdomen and breast are reviewed. A companion review paper (O'Shea et al 2015 Phys. Med. Biol. submitted) will extensively discuss the use of US imaging for intra-fraction motion quantification and novel applications of US technology to RT.

Multiparametric-MRI and Targeted Biopsies in the Management of Prostate Cancer Patients on Active Surveillance

An important key to clinical management of prostate cancer patients is to determine early those who will benefit from primary treatment and are not good candidates for active surveillance (AS). We describe a 67-year-old gentleman with a long history of stable prostate-specific antigen (PSA) levels and a negative biopsy. After slight PSA rise and low volume Gleason score 6 biopsy, the patient was considered for primary treatment or AS. A multiparametric (MP)-MRI exam revealed a suspicious lesion in the anterior apex of the prostate. Biopsies were carried out on a 3D-ultrasound prostate biopsy system with MRI-fusion. The location of the target area was challenging and could have been missed using standard 12-core biopsy template. The pathology determined Gleason 3 + 4 disease in 30% of the core from this region. Consequently, the patient underwent radiotherapy (RT). MP-MRI was also used to follow the changes from pre- to post-RT.

A homotopy-based sparse representation for fast and accurate shape prior modeling in liver surgical planning

Shape prior plays an important role in accurate and robust liver segmentation. However, liver shapes have complex variations and accurate modeling of liver shapes is challenging. Using large-scale training data can improve the accuracy but it limits the computational efficiency. In order to obtain accurate liver shape priors without sacrificing the efficiency when dealing with large-scale training data, we investigate effective and scalable shape prior modeling method that is more applicable in clinical liver surgical planning system. We employed the Sparse Shape Composition (SSC) to represent liver shapes by an optimized sparse combination of shapes in the repository, without any assumptions on parametric distributions of liver shapes. To leverage large-scale training data and improve the computational efficiency of SSC, we also introduced a homotopy-based method to quickly solve the L1-norm optimization problem in SSC. This method takes advantage of the sparsity of shape modeling, and solves the original optimization problem in SSC by continuously transforming it into a series of simplified problems whose solution is fast to compute. When new training shapes arrive gradually, the homotopy strategy updates the optimal solution on the fly and avoids re-computing it from scratch. Experiments showed that SSC had a high accuracy and efficiency in dealing with complex liver shape variations, excluding gross errors and preserving local details on the input liver shape. The homotopy-based SSC had a high computational efficiency, and its runtime increased very slowly when repository's capacity and vertex number rose to a large degree. When repository's capacity was 10,000, with 2000 vertices on each shape, homotopy method cost merely about 11.29 s to solve the optimization problem in SSC, nearly 2000 times faster than interior point method. The dice similarity coefficient (DSC), average symmetric surface distance (ASD), and maximum symmetric surface distance measurement was 94.31 ± 3.04%, 1.12 ± 0.69 mm and 3.65 ± 1.40 mm respectively.

Skull-stripping with machine learning deformable organisms

BACKGROUND

Segmentation methods for medical images may not generalize well to new data sets or new tasks, hampering their utility. We attempt to remedy these issues using deformable organisms to create an easily customizable segmentation plan. We validate our framework by creating a plan to locate the brain in 3D magnetic resonance images of the head (skull-stripping).

NEW METHOD

Our method borrows ideas from artificial life to govern a set of deformable models. We use control processes such as sensing, proactive planning, reactive behavior, and knowledge representation to segment an image. The image may have landmarks and features specific to that dataset; these may be easily incorporated into the plan. In addition, we use a machine learning method to make our segmentation more accurate.

RESULTS

Our method had the least Hausdorff distance error, but included slightly less brain voxels (false negatives). It also had the lowest false positive error and performed on par to skull-stripping specific method on other metrics.

COMPARISON WITH EXISTING METHOD(S)

We tested our method on 838 T1-weighted images, evaluating results using distance and overlap error metrics based on expert gold standard segmentations. We evaluated the results before and after the learning step to quantify its benefit; we also compare our results to three other widely used methods: BSE, BET, and the Hybrid Watershed algorithm.

CONCLUSIONS

Our framework captures diverse categories of information needed for brain segmentation and will provide a foundation for tackling a wealth of segmentation problems.

Multi-contrast multi-atlas parcellation of diffusion tensor imaging of the human brain

In this paper, we propose a novel method for parcellating the human brain into 193 anatomical structures based on diffusion tensor images (DTIs). This was accomplished in the setting of multi-contrast diffeomorphic likelihood fusion using multiple DTI atlases. DTI images are modeled as high dimensional fields, with each voxel exhibiting a vector valued feature comprising of mean diffusivity (MD), fractional anisotropy (FA), and fiber angle. For each structure, the probability distribution of each element in the feature vector is modeled as a mixture of Gaussians, the parameters of which are estimated from the labeled atlases. The structure-specific feature vector is then used to parcellate the test image. For each atlas, a likelihood is iteratively computed based on the structure-specific vector feature. The likelihoods from multiple atlases are then fused. The updating and fusing of the likelihoods is achieved based on the expectation-maximization (EM) algorithm for maximum a posteriori (MAP) estimation problems. We first demonstrate the performance of the algorithm by examining the parcellation accuracy of 18 structures from 25 subjects with a varying degree of structural abnormality. Dice values ranging 0.8-0.9 were obtained. In addition, strong correlation was found between the volume size of the automated and the manual parcellation. Then, we present scan-rescan reproducibility based on another dataset of 16 DTI images - an average of 3.73%, 1.91%, and 1.79% for volume, mean FA, and mean MD respectively. Finally, the range of anatomical variability in the normal population was quantified for each structure.

COVARIANCE SHRINKING IN ACTIVE SHAPE MODELS WITH APPLICATION TO GYRAL LABELING OF THE CEREBRAL CORTEX

Active shape models (ASMs) have been widely used in segmentation tasks in medical image analysis. Complex structures and a limited number of training samples can, however, result in the failure to capture the complete range of shape variations. Various modifications to the point distribution model (PDM) have been proposed to increase the flexibility of the model. Still model parameters are often determined empirically without respect to the underlying data structure. We explore shrinkage covariance estimation in building a PDM by combining the sample covariance matrix with a target covariance matrix estimated from a low-dimensional constrained model. Instead of using a global shrinkage intensity, we apply a spatially varying shrinkage intensity field to better adapt to the spatially varying characteristic of a complex shape. The parameters of the constrained model and the amount of shrinkage are determined in a data-driven fashion, so that the resulting distribution is optimized in representing the underlying data. The PDM, which we call SC-PDM, shows an increased flexibility in fitting new shapes and at the same time, is robust to noise. We demonstrate the effectiveness of using SC-PDM to label gyral regions on the human cerebral cortex.

A new segmentation framework based on sparse shape composition in liver surgery planning system

PURPOSE

To improve the accuracy and the robustness of the segmentation in living donor liver transplantation (LDLT) surgery planning system, the authors present a new segmentation framework that addresses challenges induced by the complex shape variations of patients' livers with cancer. It is designed to achieve the accurate and robust segmentation of hepatic parenchyma, portal veins, hepatic veins, and tumors in the LDLT surgery planning system.

METHODS

The segmentation framework proposed in this paper includes two important modules: (1) The robust shape prior modeling for liver, in which the sparse shape composition (SSC) model is employed to deal with the complex variations of liver shapes and obtain patient-specific liver shape priors. (2) The integration of the liver shape prior with a minimally supervised segmentation algorithm to achieve the accurate segmentation of hepatic parenchyma, portal veins, hepatic veins, and tumors simultaneously. The authors apply this segmentation framework to our previously developed LDLT surgery planning system to enhance its accuracy and robustness when dealing with complex cases of patients with liver cancer.

RESULTS

Compared with the principal component analysis, the SSC model shows a great advantage in handling the complex variations of liver shapes. It also effectively excludes gross errors and outliers that appear in the input shape and preserves local details for specific patients. The proposed segmentation framework was evaluated on the clinical image data of liver cancer patients, and the average symmetric surface distance for hepatic parenchyma, portal veins, hepatic veins, and tumors was 1.07 ± 0.76, 1.09 ± 0.28, 0.92 ± 0.35 and 1.13 ± 0.37 mm, respectively. The Hausdorff distance for these four tissues was 7.68, 4.67, 4.09, and 5.36 mm, respectively.

CONCLUSIONS

The proposed segmentation framework improves the robustness of the LDLT surgery planning system remarkably when dealing with complex clinical liver shapes. The SSC model is able to handle non-Gaussian errors and preserve local detail information of the input liver shape. As a result, the proposed framework effectively addresses the problems caused by the complex shape variations of livers with cancer. Our framework not only obtains accurate segmentation results for healthy persons and common patients, but also shows high robustness when dealing with specific patients with large variations of liver shapes in complex clinical environments.

Bayesian Parameter Estimation and Segmentation in the Multi-Atlas Random Orbit Model

This paper examines the multiple atlas random diffeomorphic orbit model in Computational Anatomy (CA) for parameter estimation and segmentation of subcortical and ventricular neuroanatomy in magnetic resonance imagery. We assume that there exist multiple magnetic resonance image (MRI) atlases, each atlas containing a collection of locally-defined charts in the brain generated via manual delineation of the structures of interest. We focus on maximum a posteriori estimation of high dimensional segmentations of MR within the class of generative models representing the observed MRI as a conditionally Gaussian random field, conditioned on the atlas charts and the diffeomorphic change of coordinates of each chart that generates it. The charts and their diffeomorphic correspondences are unknown and viewed as latent or hidden variables. We demonstrate that the expectation-maximization (EM) algorithm arises naturally, yielding the likelihood-fusion equation which the a posteriori estimator of the segmentation labels maximizes. The likelihoods being fused are modeled as conditionally Gaussian random fields with mean fields a function of each atlas chart under its diffeomorphic change of coordinates onto the target. The conditional-mean in the EM algorithm specifies the convex weights with which the chart-specific likelihoods are fused. The multiple atlases with the associated convex weights imply that the posterior distribution is a multi-modal representation of the measured MRI. Segmentation results for subcortical and ventricular structures of subjects, within populations of demented subjects, are demonstrated, including the use of multiple atlases across multiple diseased groups.

Atlas-based neuroinformatics via MRI: harnessing information from past clinical cases and quantitative image analysis for patient care

With the ever-increasing amount of anatomical information radiologists have to evaluate for routine diagnoses, computational support that facilitates more efficient education and clinical decision making is highly desired. Despite the rapid progress of image analysis technologies for magnetic resonance imaging of the human brain, these methods have not been widely adopted for clinical diagnoses. To bring computational support into the clinical arena, we need to understand the decision-making process employed by well-trained clinicians and develop tools to simulate that process. In this review, we discuss the potential of atlas-based clinical neuroinformatics, which consists of annotated databases of anatomical measurements grouped according to their morphometric phenotypes and coupled with the clinical informatics upon which their diagnostic groupings are based. As these are indexed via parametric representations, we can use image retrieval tools to search for phenotypes along with their clinical metadata. The review covers the current technology, preliminary data, and future directions of this field.

Learning statistical correlation for fast prostate registration in image-guided radiotherapy

PURPOSE

In adaptive radiation therapy of prostate cancer, fast and accurate registration between the planning image and treatment images of the patient is of essential importance. With the authors' recently developed deformable surface model, prostate boundaries in each treatment image can be rapidly segmented and their correspondences (or relative deformations) to the prostate boundaries in the planning image are also established automatically. However, the dense correspondences on the nonboundary regions, which are important especially for transforming the treatment plan designed in the planning image space to each treatment image space, are remained unresolved. This paper presents a novel approach to learn the statistical correlation between deformations of prostate boundary and nonboundary regions, for rapidly estimating deformations of the nonboundary regions when given the deformations of the prostate boundary at a new treatment image.

METHODS

The main contributions of the proposed method lie in the following aspects. First, the statistical deformation correlation will be learned from both current patient and other training patients, and further updated adaptively during the radiotherapy. Specifically, in the initial treatment stage when the number of treatment images collected from the current patient is small, the statistical deformation correlation is mainly learned from other training patients. As more treatment images are collected from the current patient, the patient-specific information will play a more important role in learning patient-specific statistical deformation correlation to effectively reflect prostate deformation of the current patient during the treatment. Eventually, only the patient-specific statistical deformation correlation is used to estimate dense correspondences when a sufficient number of treatment images have been acquired from the current patient. Second, the statistical deformation correlation will be learned by using a multiple linear regression (MLR) model, i.e., ridge regression (RR) model, which has the best prediction accuracy than other MLR models such as canonical correlation analysis (CCA) and principal component regression (PCR).

RESULTS

To demonstrate the performance of the proposed method, we first evaluate its registration accuracy by comparing the deformation field predicted by our method with the deformation field estimated by the thin plate spline (TPS) based correspondence interpolation method on 306 serial prostate CT images of 24 patients. The average predictive error on the voxels around 5 mm of prostate boundary is 0.38 mm for our method of RR-based correlation model. Also, the corresponding maximum error is 2.89 mm. We then compare the speed for deformation interpolation by different methods. When considering the larger region of interest (ROI) with the size of 512 × 512 × 61, our method takes 24.41 seconds to interpolate the dense deformation field while TPS method needs 6.7 minutes; when considering a small ROI (surrounding prostate) with size of 112 × 110 × 93, our method takes 1.80 seconds, while TPS method needs 25 seconds.

CONCLUSIONS

Experimental results show that the proposed method can achieve much faster registration speed yet with comparable registration accuracy, compared to the TPS-based correspondence (or deformation) interpolation approach.

Localized component analysis for arthritis detection in the trapeziometacarpal joint

The trapeziometacarpal joint enables the prehensile function of the thumb. Unfortunately, this joint is vulnerable to osteoarthritis (OA) that typically affects the local shape of the trapezium. A novel, local statistical shape model is defined that employs a differentiable locality measure based on the weighted variance of point coordinates per mode. The simplicity of the function and the smooth derivative enable to quickly determine localized components for densely sampled surfaces. The method is employed to assess a set of 60 trapezia (38 healthy, 22 with OA). The localized components predominantly model regions affected by OA, contrary to shape variations found with PCA. Furthermore, identification of pathological trapezia based on the localized modes of variation is improved compared to PCA.

DRAMMS: Deformable registration via attribute matching and mutual-saliency weighting

A general-purpose deformable registration algorithm referred to as "DRAMMS" is presented in this paper. DRAMMS bridges the gap between the traditional voxel-wise methods and landmark/feature-based methods with primarily two contributions. First, DRAMMS renders each voxel relatively distinctively identifiable by a rich set of attributes, therefore largely reducing matching ambiguities. In particular, a set of multi-scale and multi-orientation Gabor attributes are extracted and the optimal components are selected, so that they form a highly distinctive morphological signature reflecting the anatomical and geometric context around each voxel. Moreover, the way in which the optimal Gabor attributes are constructed is independent of the underlying image modalities or contents, which renders DRAMMS generally applicable to diverse registration tasks. A second contribution of DRAMMS is that it modulates the registration by assigning higher weights to those voxels having higher ability to establish unique (hence reliable) correspondences across images, therefore reducing the negative impact of those regions that are less capable of finding correspondences (such as outlier regions). A continuously-valued weighting function named "mutual-saliency" is developed to reflect the matching uniqueness between a pair of voxels implied by the tentative transformation. As a result, voxels do not contribute equally as in most voxel-wise methods, nor in isolation as in landmark/feature-based methods. Instead, they contribute according to the continuously-valued mutual-saliency map, which dynamically evolves during the registration process. Experiments in simulated images, inter-subject images, single-/multi-modality images, from brain, heart, and prostate have demonstrated the general applicability and the accuracy of DRAMMS.

Brain MRI segmentation with multiphase minimal partitioning: a comparative study

This paper presents the implementation and quantitative evaluation of a multiphase three-dimensional deformable model in a level set framework for automated segmentation of brain MRIs. The segmentation algorithm performs an optimal partitioning of three-dimensional data based on homogeneity measures that naturally evolves to the extraction of different tissue types in the brain. Random seed initialization was used to minimize the sensitivity of the method to initial conditions while avoiding the need for a priori information. This random initialization ensures robustness of the method with respect to the initialization and the minimization set up. Postprocessing corrections with morphological operators were applied to refine the details of the global segmentation method. A clinical study was performed on a database of 10 adult brain MRI volumes to compare the level set segmentation to three other methods: “idealized” intensity thresholding, fuzzy connectedness, and an expectation maximization classification using hidden Markov random fields. Quantitative evaluation of segmentation accuracy was performed with comparison to manual segmentation computing true positive and false positive volume fractions. A statistical comparison of the segmentation methods was performed through a Wilcoxon analysis of these error rates and results showed very high quality and stability of the multiphase three-dimensional level set method.

Automatic model-guided segmentation of the human brain ventricular system from CT images

RATIONALE AND OBJECTIVES

Accurate segmentation of the brain ventricular system on computed tomographic (CT) imaging is useful in neurodiagnosis and neurosurgery. Manual segmentation is time consuming, usually not reproducible, and subjective. Because of image noise, low contrast between soft tissues, large interslice distance, large shape, and size variations of the ventricular system, no automatic method is presently available. The authors propose a model-guided method for the automated segmentation of the ventricular system.

MATERIALS AND METHODS

Fifty CT scans of patients with strokes at different sites were collected for this study. Given a brain CT image, its ventricular system was segmented in five steps: (1) a predefined volumetric model was registered (or deformed) onto the image; (2) according to the deformed model, eight regions of interest were automatically specified; (3) the intensity threshold of cerebrospinal fluid was calculated in a region of interest and used to segment all regions of cerebrospinal fluid from the entire brain volume; (4) each ventricle was segmented in its specified region of interest; and (5) intraventricular calcification regions were identified to refine the ventricular segmentation.

RESULTS

Compared to ground truths provided by experts, the segmentation results of this method achieved an average overlap ratio of 85% for the entire ventricular system. On a desktop personal computer with a dual-core central processing unit running at 2.13 GHz, about 10 seconds were required to analyze each data set.

CONCLUSION

Experiments with clinical CT images showed that the proposed method can generate acceptable results in the presence of image noise, large shape, and size variations of the ventricular system, and therefore it is potentially useful for the quantitative interpretation of CT images in neurodiagnosis and neurosurgery.

Ventricle Boundary in CT: Partial Volume Effect and Local Thresholds

We present a mathematical frame to carry out segmentation of cerebrospinal fluid (CSF) of ventricular region in computed tomography (CT) images in the presence of partial volume effect (PVE). First, the image histogram is fitted using the Gaussian mixture model (GMM). Analyzing the GMM, we find global threshold based on parameters of distributions for CSF, and for the combined white and grey matter (WGM). The parameters of distribution of PVE pixels on the boundary of ventricles are estimated by using a convolution operator. These parameters are used to calculate local thresholds for boundary pixels by the analysis of contribution of the neighbor pixels intensities into a PVE pixel. The method works even in the case of an almost unimodal histogram; it can be useful to analyze the parameters of PVE in the ground truth provided by the expert.

Computerized characterization of prostate cancer by fractal analysis in MR images

PURPOSE

To explore the potential of computerized characterization of prostate MR images by extracting the fractal features of texture and intensity distributions as indices in the differential diagnosis of prostate cancer.

MATERIALS AND METHODS

MR T2-weighted images (T2WI) of 55 patients with pathologic results detected by ultrasound guided biopsy were collected and then divided in two groups, 27 with prostate cancer (PCa) and 28 with no histological abnormality. Texture fractal dimension (TFD) and histogram fractal dimension (HFD) were calculated to analyze complexity features of regions of Interest (ROIs) selected from the peripheral zone. Two-sample t-tests were performed to evaluate group differences for both parameters. Receiver operating characteristic (ROC) analysis was used to estimate the performance of TFD and HFD for discriminating PCa.

RESULTS

Significant differences were found in both TFD and HFD between the two patient groups. The areas under the ROC curves of TFD and HFD were 0.691 and 0.966, respectively, in distinguishing prostatic carcinoma from normal peripheral zone. As characterized by the fractal indices, cancerous prostatic tissue exhibited smoother texture and lower variation in intensity distribution than normal prostatic tissue.

CONCLUSION

The study suggests that TFD and HFD depict the changes in texture and intensity distribution associated with prostate cancer on T2WI. Both TFD and HFD provide promising quantitative indices for cancer identification. HFD performs better than TFD offering a more robust MR-based indicator in the diagnosis of prostatic carcinoma.

Sampling the spatial patterns of cancer: optimized biopsy procedures for estimating prostate cancer volume and Gleason Score

Prostate biopsy is the current gold-standard procedure for prostate cancer diagnosis. Existing prostate biopsy procedures have been mostly focusing on detecting cancer presence. However, they often ignore the potential use of biopsy to estimate cancer volume (CV) and Gleason Score (GS, a cancer grade descriptor), the two surrogate markers for cancer aggressiveness and the two crucial factors for treatment planning. To fill up this vacancy, this paper assumes and demonstrates that, by optimally sampling the spatial patterns of cancer, biopsy procedures can be specifically designed for estimating CV and GS. Our approach combines image analysis and machine learning tools in an atlas-based population study that consists of three steps. First, the spatial distributions of cancer in a patient population are learned, by constructing statistical atlases from histological images of prostate specimens with known cancer ground truths. Then, the optimal biopsy locations are determined in a feature selection formulation, so that biopsy outcomes (either cancer presence or absence) at those locations could be used to differentiate, at the best rate, between the existing specimens having different (high vs. low) CV/GS values. Finally, the optimized biopsy locations are utilized to estimate whether a new-coming prostate cancer patient has high or low CV/GS values, based on a binary classification formulation. The estimation accuracy and the generalization ability are evaluated by the classification rates and the associated receiver-operating-characteristic (ROC) curves in cross validations. The optimized biopsy procedures are also designed to be robust to the almost inevitable needle displacement errors in clinical practice, and are found to be robust to variations in the optimization parameters as well as the training populations.

Automatic segmentation of the human brain ventricles from MR images by knowledge-based region growing and trimming

Automatic segmentation of the human brain ventricular system from MR images is useful in studies of brain anatomy and its diseases. Existing intensity-based segmentation methods are adaptive to large shape and size variations of the ventricular system, but may leak to the non-ventricular regions due to the non-homogeneity, noise and partial volume effect in the images. Deformable model-based methods are more robust to noise and alleviate the leakage problem, but may generate wrong results when the shape or size of the ventricle to be segmented in the images has a large difference in comparison to its model. In this paper, we propose a knowledge-based region growing and trimming approach where: (1) a model of a ventricular system is used to define regions of interest (ROI) for the four ventricles (i.e., left, right, third and fourth); (2) to segment a ventricle in its ROI, a region growing procedure is first applied to obtain a connected region that contains the ventricle, and (3) a region trimming procedure is then employed to trim the non-ventricle regions. A hysteretic thresholding is developed for the region growing procedure to cope with the partial volume effect and minimize non-ventricular regions. The domain knowledge on the shape and intensity features of the ventricular system is used for the region trimming procedure. Due to the joint use of the model-based and intensity-based approaches, our method is robust to noise and large shape and size variations. Experiments on 18 simulated and 58 clinical MR images show that the proposed approach is able to segment the ventricular system accurately with the dice similarity coefficient ranging from 91% to 99%.

A three-dimensional histological atlas of the human basal ganglia. II. Atlas deformation strategy and evaluation in deep brain stimulation for Parkinson disease

OBJECT

The localization of any given target in the brain has become a challenging issue because of the increased use of deep brain stimulation to treat Parkinson disease, dystonia, and nonmotor diseases (for example, Tourette syndrome, obsessive compulsive disorders, and depression). The aim of this study was to develop an automated method of adapting an atlas of the human basal ganglia to the brains of individual patients.

METHODS

Magnetic resonance images of the brain specimen were obtained before extraction from the skull and histological processing. Adaptation of the atlas to individual patient anatomy was performed by reshaping the atlas MR images to the images obtained in the individual patient using a hierarchical registration applied to a region of interest centered on the basal ganglia, and then applying the reshaping matrix to the atlas surfaces.

RESULTS

Results were evaluated by direct visual inspection of the structures visible on MR images and atlas anatomy, by comparison with electrophysiological intraoperative data, and with previous atlas studies in patients with Parkinson disease. The method was both robust and accurate, never failing to provide an anatomically reliable atlas to patient registration. The registration obtained did not exceed a 1-mm mismatch with the electrophysiological signatures in the region of the subthalamic nucleus.

CONCLUSIONS

This registration method applied to the basal ganglia atlas forms a powerful and reliable method for determining deep brain stimulation targets within the basal ganglia of individual patients.

Automated segmentation of the liver from 3D CT images using probabilistic atlas and multilevel statistical shape model

RATIONALE AND OBJECTIVES

An atlas-based automated liver segmentation method from three-dimensional computed tomographic (3D CT) images has been developed. The method uses two types of atlases, a probabilistic atlas (PA) and a statistical shape model (SSM).

MATERIALS AND METHODS

Voxel-based segmentation with a PA is first performed to obtain a liver region, then the obtained region is used as the initial region for subsequent SSM fitting to 3D CT images. To improve reconstruction accuracy, particularly for highly deformed livers, we use a multilevel SSM (ML-SSM). In ML-SSM, the entire shape is divided into patches, with principal component analysis applied to each patch. To avoid inconsistency among patches, we introduce a new constraint called the "adhesiveness constraint" for overlapping regions among patches.

RESULTS

The PA and ML-SSM were constructed from 20 training datasets. We applied the proposed method to eight evaluation datasets. On average, volumetric overlap of 89.2 +/- 1.4% and average distance of 1.36 +/- 0.19 mm were obtained.

CONCLUSIONS

The proposed method was shown to improve segmentation accuracy for datasets including highly deformed livers. We demonstrated that segmentation accuracy is improved using the initial region obtained with PA and the introduced constraint for ML-SSM.

Methods of artificial enlargement of the training set for statistical shape models

Due to the small size of training sets, statistical shape models often over-constrain the deformation in medical image segmentation. Hence, artificial enlargement of the training set has been proposed as a solution for the problem to increase the flexibility of the models. In this paper, different methods were evaluated to artificially enlarge a training set. Furthermore, the objectives were to study the effects of the size of the training set, to estimate the optimal number of deformation modes, to study the effects of different error sources, and to compare different deformation methods. The study was performed for a cardiac shape model consisting of ventricles, atria, and epicardium, and built from magnetic resonance (MR) volume images of 25 subjects. Both shape modeling and image segmentation accuracies were studied. The objectives were reached by utilizing different training sets and datasets, and two deformation methods. The evaluation proved that artificial enlargement of the training set improves both the modeling and segmentation accuracy. All but one enlargement techniques gave statistically significantly (p < 0.05) better segmentation results than the standard method without enlargement. The two best enlargement techniques were the nonrigid movement technique and the technique that combines principal component analysis (PCA) and finite element model (FEM). The optimal number of deformation modes was found to be near 100 modes in our application. The active shape model segmentation gave better segmentation accuracy than the one based on the simulated annealing optimization of the model weights.

Adaptation of a 3D prostate cancer atlas for transrectal ultrasound guided target-specific biopsy

Due to lack of imaging modalities to identify prostate cancer in vivo, current TRUS guided prostate biopsies are taken randomly. Consequently, many important cancers are missed during initial biopsies. The purpose of this study was to determine the potential clinical utility of a high-speed registration algorithm for a 3D prostate cancer atlas. This 3D prostate cancer atlas provides voxel-level likelihood of cancer and optimized biopsy locations on a template space (Zhan et al 2007). The atlas was constructed from 158 expert annotated, 3D reconstructed radical prostatectomy specimens outlined for cancers (Shen et al 2004). For successful clinical implementation, the prostate atlas needs to be registered to each patient's TRUS image with high registration accuracy in a time-efficient manner. This is implemented in a two-step procedure, the segmentation of the prostate gland from a patient's TRUS image followed by the registration of the prostate atlas. We have developed a fast registration algorithm suitable for clinical applications of this prostate cancer atlas. The registration algorithm was implemented on a graphical processing unit (GPU) to meet the critical processing speed requirements for atlas guided biopsy. A color overlay of the atlas superposed on the TRUS image was presented to help pick statistically likely regions known to harbor cancer. We validated our fast registration algorithm using computer simulations of two optimized 7- and 12-core biopsy protocols to maximize the overall detection rate. Using a GPU, patient's TRUS image segmentation and atlas registration took less than 12 s. The prostate cancer atlas guided 7- and 12-core biopsy protocols had cancer detection rates of 84.81% and 89.87% respectively when validated on the same set of data. Whereas the sextant biopsy approach without the utility of 3D cancer atlas detected only 70.5% of the cancers using the same histology data. We estimate 10-20% increase in prostate cancer detection rates when TRUS guided biopsies are assisted by the 3D prostate cancer atlas compared to the current standard of care. The fast registration algorithm we have developed can easily be adapted for clinical applications for the improved diagnosis of prostate cancer.

Advances in neuroimaging in patients with epilepsy

Intractable seizures can have a devastating effect on the development of a child. In children with intractable epilepsy that is refractory to medication, surgical treatment may be needed. Magnetic resonance imaging is an essential neuroimaging tool to assist in the identification of an epileptogenic substrate. The interpretation of MR images should be done in the context of clinical knowledge of the seizure symptomatology and electroencephalographic findings. Quantitative processing of structural MR data and advanced MR imaging such as diffusion tensor imaging and MR spectroscopy have the potential to identify subtle lesions that may otherwise have been missed. In addition to lesion localization, identification of eloquent cortex and white matter tracts are also an essential component of epilepsy surgery workup. Functional MR imaging maps the sensorimotor cortex and also lateralizes language. Diffusion tensor imaging tractography can be used to map the corticospinal tracts and the optic radiations. In addition to MR imaging, magnetoencephalography and nuclear medicine studies such as PET and SPECT scanning may be used to lateralize seizure focus when clinical, electrophysiological, and structural MR imaging findings are discordant.

A model-based, semi-global segmentation approach for automatic 3-D point landmark localization in neuroimages

The existing differential approaches for localization of 3-D anatomic point landmarks in 3-D images are sensitive to noise and usually extract numerous spurious landmarks. The parametric model-based approaches are not practically usable for localization of landmarks that can not be modeled by simple parametric forms. Some dedicated methods using anatomic knowledge to identify particular landmarks are not general enough to cope with other landmarks. In this paper, we propose a model-based, semi-global segmentation approach to automatically localize 3-D point landmarks in neuroimages. To localize a landmark, the semi-global segmentation (meaning the segmentation of a part of the studied structure in a certain neighborhood of the landmark) is first achieved by an active surface model, and then the landmark is localized by analyzing the segmented part only. The joint use of global model-to-image registration, semi-global structure registration, active surface-based segmentation, and point-anchored surface registration makes our method robust to noise and shape variation. To evaluate the method, we apply it to the localization of ventricular landmarks including curvature extrema, centerline intersections, and terminal points. Experiments with 48 clinical and 18 simulated magnetic resonance (MR) volumetric images show that the proposed approach is able to localize these landmarks with an average accuracy of 1 mm (i.e., at the level of image resolution). We also illustrate the use of the proposed approach to cortical landmark identification and discuss its potential applications ranging from computer-aided radiology and surgery to atlas registration with scans.

Efficient and generalizable statistical models of shape and appearance for analysis of cardiac MRI

We present a framework for the analysis of short axis cardiac MRI, using statistical models of shape and appearance. The framework integrates temporal and structural constraints and avoids common optimization problems inherent in such high dimensional models. The first contribution is the introduction of an algorithm for fitting 3D active appearance models (AAMs) on short axis cardiac MRI. We observe a 44-fold increase in fitting speed and a segmentation accuracy that is on par with Gauss-Newton optimization, one of the most widely used optimization algorithms for such problems. The second contribution involves an investigation on hierarchical 2D+time active shape models (ASMs), that integrate temporal constraints and simultaneously improve the 3D AAM based segmentation. We obtain encouraging results (endocardial/epicardial error 1.43+/-0.49 mm/1.51+/-0.48 mm) on 7980 short axis cardiac MR images acquired from 33 subjects. We have placed our dataset online, for the community to use and build upon.

Registering histologic and MR images of prostate for image-based cancer detection

RATIONALE AND OBJECTIVES

Needle biopsy is currently the only way to confirm prostate cancer. To increase prostate cancer diagnostic rate, needles are expected to be deployed at suspicious cancer locations. High-contrast magnetic resonance (MR) imaging provides a powerful tool for detecting suspicious cancerous tissues. To do this, MR appearances of cancerous tissue should be characterized and learned from a sufficient number of prostate MR images with known cancer information. However, ground-truth cancer information is only available in histologic images. Therefore it is necessary to warp ground-truth cancerous regions in histological images to MR images by a registration procedure. The objective of this article is to develop a registration technique for aligning histological and MR images of the same prostate.

MATERIAL AND METHODS

Five pairs of histological and T2-weighted MR images of radical prostatectomy specimens are collected. For each pair, registration is guided by two sets of correspondences that can be reliably established on prostate boundaries and internal salient bloblike structures of histologic and MR images.

RESULTS

Our developed registration method can accurately register histologic and MR images. It yields results comparable to manual registration, in terms of landmark distance and volume overlap. It also outperforms both affine registration and boundary-guided registration methods.

CONCLUSIONS

We have developed a novel method for deformable registration of histologic and MR images of the same prostate. Besides the collection of ground-truth cancer information in MR images, the method has other potential applications. An automatic, accurate registration of histologic and MR images actually builds a bridge between in vivo anatomical information and ex vivo pathologic information, which is valuable for various clinical studies.

Inter-subject comparison of MRI knee cartilage thickness

In this paper, we present the development and application of current image processing techniques to perform MRI inter-subject comparison of knee cartilage thickness based on the registration of bone structures. Each point in the bone surface which is part of the bone-cartilage interface is assigned a cartilage thickness value. Cartilage and corresponding bone structures are segmented and their shapes interpolated to create isotropic voxels. Cartilage thicknesses are computed for each point in the bone-cartilage interfaces and transferred to the bone surfaces. Corresponding anatomic points are then computed for bone surfaces based on shape matching using 3D shape descriptors called shape contexts to register bones with affine and elastic transformations, and then perform a point to point comparison of cartilage thickness values. An alternative technique for cartilage shape interpolation using a morphing technique is also presented. The cartilage segmentation and morphing were validated visually, based on volumetric measurements of porcine knee images which cartilage volumes were measured using a water displacement method, and based on digital thickness values computed with an established technique. Shape matching using 3D shape contexts was validated visually and against manual shape matching performed by a radiologist. The reproducibility of intra- and inter-subject cartilage thickness comparisons was established, as well as the feasibility of using the proposed technique to build a mean femoral shape, cartilage thickness map, and cartilage coverage map. Results showed that the proposed technique is robust, accurate, and reproducible to perform point to point inter-subject comparison of knee cartilage thickness values.