Hierarchical Rigid Registration of Femur Surface Model Based on Anatomical Features

Existing model registration of individual bones does not have a high certainly of success due to the lack of anatomic semantic. In light of the surface anatomy and functional structure of bones, we hypothesized individual femur models would be aligned through feature points both in geometrical level and in anatomic level, and proposed a hierarchical approach for the rigid registration (HRR) of point cloud models of femur with high resolution. Firstly, a coarse registration between two simplified point cloud models was implemented based on the extraction of geometric feature points (GFPs); and then, according to the anatomic feature points (AFPs) in two level namely shape features and structure features, the fine weight-based registration was performed to achieve anatomical alignment; finally, the origin source model was automatically transformed by applying the obtained coarse matrix and fine one in sequence. Experimental results show that the hierarchical registration method can rapidly and accurately register point clouds of individual femurs, and achieves the medical semantic alignment, and provides a basic tool for the understanding and comparison of femur anatomy and structure.     

The effectiveness of gadolinium MRI to improve target delineation for radiotherapy in hepatocellular carcinoma: A comparative study of rigid image registration techniques

To achieve consistent target delineation in radiotherapy for hepatocellular carcinoma (HCC), image registration between simulation CT and diagnostic MRI was explored.Twenty patients with advanced HCC were included. The median interval between MRI and CT was 11 days. CT was obtained with shallow free breathing and MRI at exhale phase. On each CT and MRI, the liver and the gross target volume (GTV) were drawn. A rigid image registration was taken according to point information of vascular bifurcation (Method[A]) and pixel information of volume of interest only including the periphery of the liver (Method[B]) and manually drawn liver (Method[C]). In nine cases with an indefinite GTV on CT, a virtual sphere was generated at the epicenter of the GTV. The GTV from CT (V_GTV[CT]) and MRI (V_GTV[MR]) and the expanded GTV from MRI (V_+GTV[MR]) considering geometrical registration error were defined. The underestimation (uncovered V[CT] by V[MR]) and the overestimation (excessive V[MR] by V[CT]) were calculated. Through a paired T-test, the difference between image registration techniques was analyzed.For method[A], the underestimation rates of V_GTV[MR] and V_+GTV[MR] were 16.4 ± 8.9% and 3.2 ± 3.7%, and the overestimation rates were 16.6 ± 8.7% and 28.4 ± 10.3%, respectively. For V_GTV[MR] and V_+GTV[MR], the underestimation rates and overestimation rates of method[A] were better than method[C]. The underestimation rates and overestimation rates of the V_GTV[MR] were better in method[B] than method[C]. By image registration and additional margin, about 97% of HCC could be covered. Method[A] or method[B] could be recommended according to physician preference.     

Segmentation of parotid glands from registered CT and MR images

PurposeTo develop an automatic multimodal method for segmentation of parotid glands (PGs) from pre-registered computed tomography (CT) and magnetic resonance (MR) images and compare its results to the results of an existing state-of-the-art algorithm that segments PGs from CT images only.MethodsMagnetic resonance images of head and neck were registered to the accompanying CT images using two different state-of-the-art registration procedures. The reference domains of registered image pairs were divided on the complementary PG regions and backgrounds according to the manual delineation of PGs on CT images, provided by a physician. Patches of intensity values from both image modalities, centered around randomly sampled voxels from the reference domain, served as positive or negative samples in the training of the convolutional neural network (CNN) classifier. The trained CNN accepted a previously unseen (registered) image pair and classified its voxels according to the resemblance of its patches to the patches used for training. The final segmentation was refined using a graph-cut algorithm, followed by the dilate-erode operations.ResultsUsing the same image dataset, segmentation of PGs was performed using the proposed multimodal algorithm and an existing monomodal algorithm, which segments PGs from CT images only. The mean value of the achieved Dice overlapping coefficient for the proposed algorithm was 78.8%, while the corresponding mean value for the monomodal algorithm was 76.5%.ConclusionsAutomatic PG segmentation on the planning CT image can be augmented with the MR image modality, leading to an improved RT planning of head and neck cancer.     

Registration methods in radiotherapy

PurposeThe aim of this study is to present a short and comprehensive review of the methods of medical image registration, their conditions and applications in radiotherapy. A particular focus was placed on the methods of deformable image registration.MethodsTo structure and deepen the knowledge on medical image registration in radiotherapy, a medical literature analysis was made using the Google Scholar browser and the medical database of the PubMed library.ResultsChronological review of image registration methods in radiotherapy based on 34 selected articles. A particular attention was given to show: (i) potential regions of the application of different methods of registration, (ii) mathematical basis of the deformable methods and (iii) the methods of quality control for the registration process.ConclusionsThe primary aim of the medical image registration process is to connect the contents of images. What we want to achieve is a complementary or extended knowledge that can be used for more precise localisation of pathogenic lesions and continuous improvement of patient treatment. Therefore, the choice of imaging mode is dependent on the type of clinical study. It is impossible to visualise all anatomical details or functional changes using a single modality machine. Therefore, fusion of various modality images is of great clinical relevance. A natural problem in analysing the fusion of medical images is geographical errors related to displacement. The registered images are performed not at the same time and, very often, at different respiratory phases.     

Pair-wise vs group-wise registration in statistical shape model construction: representation of physiological and pathological variability of bony surface morphology

Statistical shape models (SSM) of bony surfaces have been widely proposed in orthopedics, especially for anatomical bone modeling, joint kinematic analysis, staging of morphological abnormality, and pre- and intra-operative shape reconstruction. In the SSM computation, reference shape selection, shape registration and point correspondence computation are fundamental aspects determining the quality (generality, specificity and compactness) of the SSM. Such procedures can be made critical by the presence of large morphological dissimilarities within the surfaces, not only because of anthropometrical variability but also mainly due to pathological abnormalities. In this work, we proposed a SW pipeline for SSM construction based on pair-wise (PW) shape registration, which requires the a-priori selection of the reference shape, and on a custom iterative point correspondence algorithm. We addressed large morphological deformations in five different bony surface sets, namely proximal femur, distal femur, patella, proximal fibula and proximal tibia, extracted from a retrospective patient dataset. The technique was compared to a method from the literature, based on group-wise (GW) shape registration. As a main finding, the proposed technique provided generalization and specificity median errors, for all the five bony regions, lower than 2?mm. The comparative analysis provided basically similar results. Particularly, for the distal femur that was the shape affected by the largest pathological deformations, the differences in generalization, specificity and compactness were lower than 0.5?mm, 0.5?mm, and 1%, respectively. We can argue the proposed pipeline, along with the robust correspondence algorithm, is able to compute high-quality SSM of bony shapes, even affected by large morphological variability.     

Augmenting intraoperative ultrasound with preoperative magnetic resonance planning models for percutaneous renal access

ABSTRACT: BACKGROUND: Ultrasound (US) is a commonly-used intraoperative imaging modality for guiding percutaneous renal access (PRA). However, the anatomy identification and target localization abilities of the US imaging are limited. This paper evaluates the feasibility and efficiency of a proposed image-guided PRA by augmenting the intraoperative US with preoperative magnetic resonance (MR) planning models. METHODS: First, a preoperative surgical planning approach is presented to define an optimal needle trajectory using MR volume data. Then, a MR to US registration is proposed to transfer the preoperative planning into the intraoperative context. The proposed registration makes use of orthogonal US slices to avoid local minima while reduce processing time. During the registration, a respiratory gating method is used to minimize the impact of kidney deformation. By augmenting the intraoperative US with preoperative MR models and a virtual needle, a visual guidance is provided to guarantee the correct execution of the surgical planning. The accuracy, robustness and processing time of the proposed registration were evaluated by four urologists on human data from four volunteers. Furthermore, the PRA experiments were performed by the same four urologists on a kidney phantom. The puncture accuracy in terms of the needle-target distance was measured, while the perceptual quality in using the proposed image guidance was evaluated according to custom scoring method. RESULTS: The mean registration accuracy in terms of the root mean square (RMS) target registration error (TRE) is 3.53mm. The RMA distance from the registered feature points to their average is 0.81mm. The mean operating time of the registration is 6'4". In the phantom evaluation, the mean needle-target distance is 2.08mm for the left lesion and 1.85mm for the right one. The mean duration for all phantom PRA tests was 4'26". According to the custom scoring method, the mean scores of the Intervention Improvement, Workflow Impact, and Clinical Relevance were 4.0, 3.3 and 3.9 respectively. CONCLUSIONS: The presented image guidance is feasible and promising for PRA procedure. With careful setup it can be efficient for overcoming the limitation of current US-guided PRA.     

Estimation of 3D shape, internal density and mechanics of proximal femur by combining bone mineral density images with shape and density templates

Measurement of bone mineral density (BMD) by dual-energy X-ray absorptiometry (DXA) alone is only a moderate predictor of fracture risk. Finite element analysis (FEA) of bone mechanics, based on DXA images, may improve the prediction of fracture risk. We developed a method to estimate the 3D shape and density distribution of the proximal femur, using a 2D BMD image and a femur shape template. Proximal femurs of eighteen human cadavers were imaged using computed tomography and divided into two sets (N = 9 + 9). The template was created from the samples in first set by using 3D generalized Procrustes analysis and thin-plate splines. Subsequently, the template and 2D BMD image were utilized to estimate the shape and internal density distribution of the femurs in the second set. Finally, FEA was conducted based on the original and the estimated bone models to evaluate the effect of geometrical and density distributional errors on the mechanical strength. The volumetric errors induced by the estimation itself were low (<1.4%). In the estimation of bones in the second set, the mean distance difference between the estimated and the original bone surfaces was 0.80 ± 0.19 mm, suggesting feasible estimation of the femoral shape. The mean absolute error in voxel-by-voxel BMD was 120±8 mg cm?3. In FEA, the stiffness of the proximal femur differed by -7±16% between the original and estimated bones. The present method, in comparison with methods used in previous studies, improved the prediction of the geometry, the BMD distribution and the mechanical characteristics of the proximal femur. Potentially, the proposed method could ultimately improve the determination of bone fracture risk.     

Development of a parametric finite element model of the proximal femur using statistical shape and density modelling

Skeletal fractures associated with bone mass loss are a major clinical problem and economic burden, and lead to significant morbidity and mortality in the ageing population. Clinical image-based measures of bone mass show only moderate correlative strength with bone strength. However, engineering models derived from clinical image data predict bone strength with significantly greater accuracy. Currently, image-based finite element (FE) models are time consuming to construct and are non-parametric. The goal of this study was to develop a parametric proximal femur FE model based on a statistical shape and density model (SSDM) derived from clinical image data. A small number of independent SSDM parameters described the shape and bone density distribution of a set of cadaver femurs and captured the variability affecting proximal femur FE strength predictions. Finally, a three-dimensional FE model of an 'unknown' femur was reconstructed from the SSDM with an average spatial error of 0.016 mm and an average bone density error of 0.037 g/cm(3).     

Non-rigid registration of serial dedicated breast CT,longitudinal dedicated breast CT and PET/CT images using the diffeomorphic demons method

Rationale and objectivesDedicated breast CT and PET/CT scanners provide detailed 3D anatomical and functional imaging data sets and are currently being investigated for applications in breast cancer management such as diagnosis, monitoring response to therapy and radiation therapy planning. Our objective was to evaluate the performance of the diffeomorphic demons (DD) non-rigid image registration method to spatially align 3D serial (pre- and post-contrast) dedicated breast computed tomography (CT), and longitudinally-acquired dedicated 3D breast CT and positron emission tomography (PET)/CT images.MethodsThe algorithmic parameters of the DD method were optimized for the alignment of dedicated breast CT images using training data and fixed. The performance of the method for image alignment was quantitatively evaluated using three separate data sets; (1) serial breast CT pre- and post-contrast images of 20 women, (2) breast CT images of 20 women acquired before and after repositioning the subject on the scanner, and (3) dedicated breast PET/CT images of 7 women undergoing neo-adjuvant chemotherapy acquired pre-treatment and after 1 cycle of therapy.ResultsThe DD registration method outperformed no registration (p < 0.001) and conventional affine registration (p ≤ 0.002) for serial and longitudinal breast CT and PET/CT image alignment. In spite of the large size of the imaging data, the computational cost of the DD method was found to be reasonable (3–5 min).ConclusionsCo-registration of dedicated breast CT and PET/CT images can be performed rapidly and reliably using the DD method. This is the first study evaluating the DD registration method for the alignment of dedicated breast CT and PET/CT images.     

A methodology for registration of a histological slide and in vivo MRI volume based on optimizing mutual information

We present a method for registering histology and in vivo imaging that requires minimal microtoming and is automatic following the user's initialization. In this demonstration, we register a single hematoxylin-and-eosin-stained histological slide of a coronal section of a rat brain harboring a 9L gliosarcoma with an in vivo 7T MR image volume of the same brain. Because the spatial resolution of the in vivo MRI is limited, we add the step of obtaining a high spatial resolution, ex vivo MRI in situ for intermediate registration. The approach taken was to maximize mutual information in order to optimize the registration between all pairings of image data whether the sources are MRI, tissue block photograph, or stained sample photograph. The warping interpolant used was thin plate splines with the appropriate basis function for either 2-D or 3-D applications. All registrations were implemented by user initialization of the approximate pose between the two data sets, followed by automatic optimization based on maximizing mutual information. Only the higher quality anatomical images were used in the registration process; however, the spatial transformation was directly applied to a quantitative diffusion image. Quantitative diffusion maps from the registered location appeared highly correlated with the H&E slide. Overall, this approach provides a robust method for coregistration of in vivo images with histological sections and will have broad applications in the field of functional and molecular imaging.     

基于非刚体配准的体数据插值重建方法

三维图像的处理和操作需要将一般的断层序列插值成为具有各坐标轴一致的分辨率的体数据,而目前最常用的线性插值方法在层间距较大时会导致图像边缘模糊和出现伪影。Penney根据现有的非刚体匹配方法,提出了利用图像形变场数据的插值算法,大大提高了层间插值的质量。本文对Penney提出的算法进行了两方面的改进,在配准过程中用简单的单射性约束取代了复杂的平滑性约束,用邻域平均算法替代Penney使用的最邻近直线插值方法,并将新算法的实验结果与原算法、线性插值进行了对比,新算法在保持高质量插值的前提下提高了计算速度。该算法可以应用于精度要求比较高的体数据插值重建过程。     

Concept and development of an orthotropic FE model of the proximal femur

PURPOSE: In contrast to many isotropic finite-element (FE) models of the femur in literature, it was the object of our study to develop an orthotropic FE "model femur" to realistically simulate three-dimensional bone remodelling. METHODS: The three-dimensional geometry of the proximal femur was reconstructed by CT scans of a pair of cadaveric femurs at equal distances of 2mm. These three-dimensional CT models were implemented into an FE simulation tool. Well-known "density-determined" bony material properties (Young's modulus; Poisson's ratio; ultimate strength in pressure, tension and torsion; shear modulus) were assigned to each FE of the same "CT-density-characterized" volumetric group.In order to fix the principal directions of stiffness in FE areas with the same "density characterization", the cadaveric femurs were cut in 2mm slices in frontal (left femur) and sagittal plane (right femur). Each femoral slice was scanned into a computer-based image processing system. On these images, the principal directions of stiffness of cancellous and cortical bone were determined manually using the orientation of the trabecular structures and the Haversian system. Finally, these geometric data were matched with the "CT-density characterized" three-dimensional femur model. In addition, the time and density-dependent adaptive behaviour of bone remodelling was taken into account by implementation of Carter's criterion. RESULTS: In the constructed "model femur", each FE is characterized by the principal directions of the stiffness and the "CT-density-determined" material properties of cortical and cancellous bone. Thus, on the basis of anatomic data a three-dimensional FE simulation reference model of the proximal femur was realized considering orthotropic conditions of bone behaviour. CONCLUSIONS: With the orthotropic "model femur", the fundamental basis has been formed to realize realistic simulations of the dynamical processes of bone remodelling under different loading conditions or operative procedures (osteotomies, total hip replacements, etc).     

The use of sparse CT datasets for auto-generating accurate FE models of the femur and pelvis

The finite element (FE) method when coupled with computed tomography (CT) is a powerful tool in orthopaedic biomechanics. However, substantial data is required for patient-specific modelling. Here we present a new method for generating a FE model with a minimum amount of patient data. Our method uses high order cubic Hermite basis functions for mesh generation and least-square fits the mesh to the dataset. We have tested our method on seven patient data sets obtained from CT assisted osteodensitometry of the proximal femur. Using only 12 CT slices we generated smooth and accurate meshes of the proximal femur with a geometric root mean square (RMS) error of less than 1 mm and peak errors less than 8 mm. To model the complex geometry of the pelvis we developed a hybrid method which supplements sparse patient data with data from the visible human data set. We tested this method on three patient data sets, generating FE meshes of the pelvis using only 10 CT slices with an overall RMS error less than 3 mm. Although we have peak errors about 12 mm in these meshes, they occur relatively far from the region of interest (the acetabulum) and will have minimal effects on the performance of the model. Considering that linear meshes usually require about 70-100 pelvic CT slices (in axial mode) to generate FE models, our method has brought a significant data reduction to the automatic mesh generation step. The method, that is fully automated except for a semi-automatic bone/tissue boundary extraction part, will bring the benefits of FE methods to the clinical environment with much reduced radiation risks and data requirement.     

Possibility Study of Scale Invariant Feature Transform (SIFT) Algorithm Application to Spine Magnetic Resonance Imaging

The purpose of this study is an application of scale invariant feature transform (SIFT) algorithm to stitch the cervical-thoracic-lumbar (C-T-L) spine magnetic resonance (MR) images to provide a view of the entire spine in a single image. All MR images were acquired with fast spin echo (FSE) pulse sequence using two MR scanners (1.5 T and 3.0 T). The stitching procedures for each part of spine MR image were performed and implemented on a graphic user interface (GUI) configuration. Moreover, the stitching process is performed in two categories; manual point-to-point (mPTP) selection that performed by user specified corresponding matching points, and automated point-to-point (aPTP) selection that performed by SIFT algorithm. The stitched images using SIFT algorithm showed fine registered results and quantitatively acquired values also indicated little errors compared with commercially mounted stitching algorithm in MRI systems. Our study presented a preliminary validation of the SIFT algorithm application to MRI spine images, and the results indicated that the proposed approach can be performed well for the improvement of diagnosis. We believe that our approach can be helpful for the clinical application and extension of other medical imaging modalities for image stitching.     

Multimodality infarct identification for optimal image-guided intramyocardial cell injections

BackgroundIntramyocardial cell injections in the context of cardiac regenerative therapy can currently be performed using electromechanical mapping (EMM) provided by the NOGA®XP catheter injection system. The gold standard technique to determine infarct size and location, however, is late gadolinium enhanced magnetic resonance imaging (LGE-MRI). In this article we describe a practical and accurate technique to co-register LGE-MRI and NOGA®XP datasets during the injection procedures to ultimately perform image-guided injections to the border zone of the infarct determined by LGE-MRI.ResultsImage registration was successful in all datasets, and resulted in a mean registration error of 3.22 ± 1.86 mm between the MRI surface mesh and EMM points. Visual assessment revealed that the locations and the transmural extent of the infarctions measured by LGE-MRI only partly overlap with the infarct areas identified by the EMM parameters.ConclusionsThe 3D CartBox image registration toolbox enables registration of EMM on pre-procedurally acquired MRI during the catheter injection procedure. This allows the operator to perform real-time image-guided cell injections into the border zone of the infarct as assessed by LGE-MRI. The 3D CartBox thereby enables, for the first time, standardisation of the injection location for cardiac regenerative therapy.

Electronic supplementary material

The online version of this article (doi:10.1007/s12471-014-0604-2) contains supplementary material, which is available to authorized users.     

Facilitating tumor functional assessment by spatially relating 3D tumor histology and in vivo MRI: image registration approach

Background

Magnetic resonance imaging (MRI), together with histology, is widely used to diagnose and to monitor treatment in oncology. Spatial correspondence between these modalities provides information about the ability of MRI to characterize cancerous tissue. However, registration is complicated by deformations during pathological processing, and differences in scale and information content.

Methodology/Principal Findings

This study proposes a methodology for establishing an accurate 3D relation between histological sections and high resolution in vivo MRI tumor data. The key features of the methodology are: 1) standardized acquisition and processing, 2) use of an intermediate ex vivo MRI, 3) use of a reference cutting plane, 4) dense histological sampling, 5) elastic registration, and 6) use of complete 3D data sets. Five rat pancreatic tumors imaged by T2*-w MRI were used to evaluate the proposed methodology. The registration accuracy was assessed by root mean squared (RMS) distances between manually annotated landmark points in both modalities. After elastic registration the average RMS distance decreased from 1.4 to 0.7 mm. The intermediate ex vivo MRI and the reference cutting plane shared by all three 3D images (in vivo MRI, ex vivo MRI, and 3D histology data) were found to be crucial for the accurate co-registration between the 3D histological data set and in vivo MRI. The MR intensity in necrotic regions, as manually annotated in 3D histology, was significantly different from other histologically confirmed regions (i.e., viable and hemorrhagic). However, the viable and the hemorrhagic regions showed a large overlap in T2^*-w MRI signal intensity.

Conclusions

The established 3D correspondence between tumor histology and in vivo MRI enables extraction of MRI characteristics for histologically confirmed regions. The proposed methodology allows the creation of a tumor database of spatially registered multi-spectral MR images and multi-stained 3D histology.     

Variation of trabecular architecture in proximal femur of postmenopausal women

This investigation of microstructure in the human proximal femur probes the relationship between the parameters of the FRAX index of fracture risk and the parameters of bone microstructure. The specificity of fracture sites at the proximal femur raises the question of whether trabecular parameters are site-specific during post-menopause, before occurrence of fragility fracture. The donated proximal femurs of sixteen post-menopausal women in the sixth and seventh decades of life, free of metabolic pathologies and therapeutic interventions that could have altered the bone tissue, constituted the material of the study. We assessed bone mineral density of the proximal femurs by dual energy X-ray absorptiometry and then sectioned the femurs through the center of the femoral head and along the femoral neck axis. For each proximal femur, morphometry of trabeculae was conducted on the plane of the section divided into conventional regions and sub-regions consistent with the previously identified trabecular families that provide regions of relatively homogeneous microstructure. Mean trabecular width and percent bone area were calculated at such sites. Our findings indicate that each of mean trabecular width and percent bone area vary within each proximal femur independently from each other, with dependence on site. Both trabecular parameters show significant differences between pairs of sites. We speculate that a high FRAX index at the hip corresponds to a reduced percent bone area among sites that gives a more homogeneous and less site-specific quality to the proximal femur. This phenomenon may render the local tissue less able to carry out the expected mechanical function.     

动力加压髋螺钉对股骨上段生物力学特征性的影响

目的：探讨股骨上端骨折,以动力加压髋螺钉进行骨固定治疗,骨折愈合后,取出动力加压髋螺钉以后的股骨上段与完整的股骨上段的生物力学特性相比较,为临床内固定取出术后功能锻炼的强度提供量化依据。方法：收集8具新鲜尸体股骨标本进行实验应力分析,分别测定完整股骨上段和动力加压髋螺钉取出后股骨上段的力学特性改变。结果：动力加压髋螺钉取出术后股骨上段的力学特性与完整股骨上段的力学特性相比有显著的差异（P＜0．01）。结论：股骨上端骨折如果以动力加压髋螺钉为治疗手段,在骨折愈合取出内固定后,功能锻炼只能控制在慢速步行水平,不能进行奔跑、跳跃等活动,以防止再骨折等并发症的发生。     

Validation of a method for “dose of the day” calculation in head-neck tomotherapy by using planning ct-to-MVCT deformable image registration

PurposeThe aim of this study was to test the feasibility and dosimetric accuracy of a method that employs planning CT-to-MVCT deformable image registration (DIR) for calculation of the daily dose for head and neck (HN) patients treated with Helical Tomotherapy (HT).MethodsFor each patient, the planning kVCT (CTplan) was deformably registered to the MVCT acquired at the 15th therapy session (MV15) with a B-Spline Free Form algorithm using Mattes mutual information (open-source software 3D Slicer), resulting in a deformed CT (CTdef). On the same day as MVCT15, a kVCT was acquired with the patient in the same treatment position (CT15). The original HT plans were recalculated both on CTdef and CT15, and the corresponding dose distributions were compared; local dose differences <2% of the prescribed dose (DD2%) and 2D/3D gamma-index values (2%-2 mm) were assessed respectively with Mapcheck SNC Patient software (Sun Nuclear) and with 3D-Slicer.ResultsOn average, 87.9% ± 1.2% of voxels were found for DD2% (on average 27 slices available for each patient) and 94.6% ± 0.8% of points passed the 2D gamma analysis test while the 3D gamma test was satisfied in 94.8% ± 0.8% of body’s voxels.ConclusionsThis study represents the first demonstration of the dosimetric accuracy of kVCT-to-MVCT DIR for dose of the day computations. The suggested method is sufficiently fast and reliable to be used for daily delivered dose evaluations in clinical strategies for adaptive Tomotherapy of HN cancer.     

Rotationally resliced 3D prostate segmentation of MR images using Bhattacharyya similarity and active band theory

PurposeIn this article, we propose a novel, semi-automatic segmentation method to process 3D MR images of the prostate using the Bhattacharyya coefficient and active band theory with the goal of providing technical support for computer-aided diagnosis and surgery of the prostate.MethodsOur method consecutively segments a stack of rotationally resectioned 2D slices of a prostate MR image by assessing the similarity of the shape and intensity distribution in neighboring slices. 2D segmentation is first performed on an initial slice by manually selecting several points on the prostate boundary, after which the segmentation results are propagated consecutively to neighboring slices. A framework of iterative graph cuts is used to optimize the energy function, which contains a global term for the Bhattacharyya coefficient with the help of an auxiliary function. Our method does not require previously segmented data for training or for building statistical models, and manual intervention can be applied flexibly and intuitively, indicating the potential utility of this method in the clinic.ResultsWe tested our method on 3D T2-weighted MR images from the ISBI dataset and PROMISE12 dataset of 129 patients, and the Dice similarity coefficients were 90.34 ± 2.21% and 89.32 ± 3.08%, respectively. The comparison was performed with several state-of-the-art methods, and the results demonstrate that the proposed method is robust and accurate, achieving similar or higher accuracy than other methods without requiring training.ConclusionThe proposed algorithm for segmenting 3D MR images of the prostate is accurate, robust, and readily applicable to a clinical environment for computer-aided surgery or diagnosis.