《多分辨率水平集算法的乳腺MR圖像分割》
針對乳腺 MR 圖像信息量大、灰度不均勻、邊界模糊、難分割的特點, 提出一種多分辨率水平集乳腺 MR圖像分割算法. 算法的核心是首先利用小波多尺度分解對圖像進行多尺度空間分析, 得到粗尺度圖像; 然后對粗尺度圖像利用改進 CV 模型進行分割. 為了去除乳腺 MR 圖像中灰度偏移場對分割效果的影響, 算法中引入局部擬合項, 並用核函數進一步改進 CV模型, 進而對粗尺度分割效果進行優化處理. 仿真和臨床數據分割結果表明, 所提算法分割灰度不均勻圖像具有較高的分割精度和魯棒性, 能夠有效的實現乳腺 MR 圖像的分割。
《三維肝臟MR圖像分割技術研究》
醫學影像學發展至今，已經廣泛地應用於臨床醫學的各個相關鄰域。利用合適的圖像處理算法對醫學圖像進行相應的處理，能夠對基於醫學圖像的診斷以及其他研究工作提供更加有效、便捷的信息，醫學圖像的分割在醫學圖像處理中占據着重要的位置。從醫學影像中可通過分割算法提取出感興趣區域並予以單獨顯示，能夠更加直觀地提供病變或正常組織結構信息，並且分割的結果可以應用在為一定目的而進行的后續處理當中，比如圖像配准、目標組織的定量測量等。 磁共振成像技術在當前醫學研究與臨床診療中發揮着愈加重要的作用，與其他成像方式相比，MRI對軟組織和內臟的成像能力高，能夠非常清晰的顯示人體組織解剖結構，並具有多參數（T1、T2等）、多方位成像的優點。MR圖像的成像效果很好地區分了各個組織，在此基礎上可以對感興趣區域進行更為直觀地分割。近年來國民肝部病變的多發使得基於腹部掃描圖像的肝臟分割成為亟待解決的問題，然而人體腹部包含大量臟器及軟組織，結構復雜，並且臟器與軟組織之間的粘連導致成像結果中存在浸潤現象，從而形成大量弱邊緣和偽邊緣，這使得面向內臟的分割非常困難。再加上磁共振成像過程較為復雜，成像效果存在一定的不確定性，不同的組織器官之間廣泛存在的差異性，准確地從腹部MR掃描圖像中提取出肝臟具有重要的理論意義以及應用價值。 本文系統的分析了當前應用於醫學圖像分割的多種算法，對它們的優劣勢以及應用范圍進行了比較和總結。根據腹部圖像的特點選取水平集算法對肝臟進行提取，詳細描述了水平集算法的原理、特征，以及發展至今研究人員對其進行的各種改進和應用。由於人體結構的復雜性和個體之間的差異性，圖像分割算法發展至今仍然沒有一種單一的方法對人體各個部位達到有效的分割，當前主要的研究方向是綜合多種算法的優點，結合目標分割區域的形態特征進行混合分割。本文課題就是在此前提下分析考量了多種算法並研究了人體肝臟的形態特征和成像特點之后，選用閾值分割算法與水平集結合的方式，並加入一些其他的算法進行輔助分割，較好的實現了三維腹部圖像肝臟的提取工作。 本文主要研究工作如下： 一、首先將從醫學影像設備中獲取的序列切片圖像根據掃描間隔和切片層厚進行堆疊，為使其更加接近真實人體數據在切片間進行插值，為保證數據的真實性插值的數據盡量減小。 二、對插值后的體數據進行降噪濾波，由於水平集算法對圖像邊界信息敏感，要盡量保持圖像中的邊緣，采用高斯濾波或各向異性擴散濾波均可達到良好的效果。 三、使用閾值分割與水平集結合對肝臟進行提取，並在此步驟中加入非線性映射，在增強圖像的同時產生良好的速度圖像，使得分割結果中的演化溢出現象得以避免。 四、結合可視化工具包VTK使用光線投射算法對分割結果以及中間步驟各個算法的處理效果進行三維重建。 實驗結果表明本課題所選用的算法結合方式獲得了較為理想的分割效果，很好的將水平集算法應用到了三維肝臟的分割工作當中，有效的避免了水平集算法在弱邊緣處泄露的問題，為針對肝臟的后續研究提供了基礎。
《結合非局部均值的快速FCM算法分割MR圖像研究》
針對FCM算法分割醫學MR圖像存在的運算速度慢、對初始值敏感以及難以處理MR圖像中固有Rician噪聲等缺陷,提出了一種結合非局部均值的快速FCM算法。該算法的核心是首先針對MR圖像中存在的Rician噪聲,利用非局部均值算法對圖像進行去噪處理,消除噪聲對分割結果的影響；然后根據所提出的新的自動獲取聚類中心的規則得到初始聚類中心；最后將得到的聚類中心作為快速FCM算法的初始聚類中心用於去噪后的圖像分割,解決了隨機選擇初始聚類中心造成的搜索速度慢和容易陷入局部極值的問題。實驗表明,該算法能夠快速有效地分割圖像,並且具有較好的抗噪能力。
《MR圖像中的肝臟分割和腫瘤提取》
磁共振MR(Magnetic Resonance)圖像是公認的確認肝臟有無腫瘤等器質性病變的金標准檢查方法,其中涉及肝臟的分割以及腫瘤的提取.由於臟器組織浸潤和個體差異,在解決肝臟分割和腫瘤提取方面還沒有通用的數字圖像處理方法.在現有研究的基礎上,以迭代四叉樹(IQD)自動分割算法和基於灰度的分割方法,實現MR圖像中肝臟的自動分割和腫瘤的提取.實驗結果表明,這一套方法的可行性和優勢.
《基於圖划分的形狀統計主動輪廓模型心臟MR圖像分割》
為有效分析心臟功能，高精度分割左、右心室是必要的.心臟MR圖像中存在圖像灰度不均，左、右心室及周圍其它組織灰度接近，存在弱邊緣、邊緣斷裂及噪聲造成邊緣模糊等現象，給精確分割左、右心室輪廓帶來困難.本文在基於圖划分的主動輪廓方法基礎上，通過對訓練形狀進行配准及變化模式分析，定義左、右心室輪廓形狀變化允許空間，提出基於圖划分的形狀統計主動輪廓模型來分割心臟MR圖像.該方法通過圖划分理論將圖像分割問題轉化為最優化問題，所以能夠得到全局最優解，具有較大的捕捉范圍.還引入形狀統計來引導曲線的演化，有效處理曲線演化時存在的邊緣泄漏問題，提高分割精度.實驗結果表明，本文方法較以往方法具有更高的分割精度和更好的穩定性，為臨床應用提供一種較可行的方法.
《Cardiac MR Image Segmentation Techniques: an overview》
Broadly speaking, the objective in cardiac image segmentation is to delineate the outer and inner walls of the heart to segment out either the entire or parts of the organ boundaries. This paper will focus on MR images as they are the most widely used in cardiac segmentation – as a result of the accurate morphological information and better soft tissue contrast they provide. This cardiac segmentation information is very useful as it eases physical measurements that provides useful metrics for cardiac diagnosis such as infracted volumes, ventricular volumes, ejection fraction, myocardial mass, cardiac movement, and the like. But, this task is difficult due to the intensity and texture similarities amongst the different cardiac and background structures on top of some noisy artifacts present in MR images. Thus far, various researchers have proposed different techniques to solve some of the pressing issues. This seminar paper presents an overview of representative medical image segmentation techniques. The paper also highlights preferred approaches for segmentation of the four cardiac chambers: the left ventricle (LV), right ventricle (RV), left atrium (LA) and right atrium (RA), on short axis image planes.
《MR Image Segmentation of Left Ventricle Based on the Multi-information Gaussian Mixture Model》
The Level set method has consequence in the fields of image segmentations.As the traditional active contour methods only use the information of the edge,when it segments images with strong noise or with weak edges it is difficult to get the true edge.Gaussian mixture model uses the global information of the image,so it can do solve the problems of the weak edges.But the traditional Gaussian mixture model only uses the information of the histogram and not uses the information of the location of the pixel.So it is sensitive to the noise.This paper gives a method to make a new information field,which combines the information of the region,texture and region simulation.With the new information field the Gaussian mixture model can reduce the effect of the noise.In this paper the Gaussian mixture model is introduced to the Level set model and reduces the effect of the noise and prevents the curve over the weak edges.After getting the inner edge of the left ventricle,this paper uses the region and shape information to segment the out edge.Experiments on the segmentation of left ventricle magnetic resonance images show this model has better effect in image segmentation.
《Prostate MR image segmentation using 3D Active Appearance Models》
This paper presents a method for automatic segmentation of the prostate from transversal T2-weighted images based on 3D Active Appearance Models (AAM). The algorithm consist of two stages. Firstly, Shape Context based non-rigid surface registration of the manual segmented images is used to obtain the point correspondence between the given training cases. Subsequently, an AAM is used to segment the prostate on 50 training cases. The method is evaluated using a 5-fold cross validation over 5 repetitions. The mean Dice similarity coefficient and 95% Hausdorff distance are 0.78 and 7.32 mm respectively. Prostate segmentation is essential for calculating prostate volume, image fusion, creating patient-specific prostate anatomical models, and as a pre-processing step for many computer aided diagnosis algorithms. Furthermore, information about the size, volume, shape and location of the prostate relative to adjacent organs is an essential part of planning for minimally invasive therapies and biopsies. Because manual segmentation of the prostate is time-consuming and highly subjective, (semi-)automatic segmentation methods are preferable. However, segmenting the prostate in MR images is challenging due to the large variations of prostate shape between subjects, the lack of clear prostate boundaries and the similar intensity profiles of the prostate and surrounding tissues. The 2012 MICCAI challenge: " Prostate MR Image Segmentation " involves segmentation of the prostate on transversal T2-weighted images. The goal of the challenge is to evaluate segmentation algorithms on images from multiple centers and multiple MRI device vendors. Only a few prostate segmentation methods for T2-weighted MR images currently exist. Klein et al. [1] proposed a method based on non-rigid registration of a set of pre-labeled atlas images, against the target patients image, using mutual information. Subsequently, the segmentation is obtained as the average of the best matched registered atlas sets. Multiple modifications are published on this atlas based prostate segmentation method [2–4]. The methods presented by Toth et al. [5] and Ghose et al. [6, 7] are based on statistical shape models. Toth et al. used a levelset-based statistical shape。
《A combinatorial Bayesian and Dirichlet model for prostate MR image segmentation using probabilistic image features》
Blurred boundaries and heterogeneous intensities make accurate prostate MR image segmentation problematic. To improve prostate MR image segmentation we suggest an approach that includes: (a) an image patch division method to partition the prostate into homogeneous segments for feature extraction; (b) an image feature formulation and classification method, using the relevance vector machine, to provide probabilistic prior knowledge for graph energy construction; (c) a graph energy formulation scheme with Bayesian priors and Dirichlet graph energy and (d) a non-iterative graph energy minimization scheme, based on matrix differentiation, to perform the probabilistic pixel membership optimization. The segmentation output was obtained by assigning pixels with foreground and background labels based on derived membership probabilities. We evaluated our approach on the PROMISE-12 dataset with 50 prostate MR image volumes. Our approach achieved a mean dice similarity coefficient (DSC) of 0.90 ± 0.02, which surpassed the five best prior-based methods in the PROMISE-12 segmentation challenge.
《Fully Automatic Localization and Segmentation of 3D Vertebral Bodies from CT/MR Images via a Learning-Based Method》
In this paper, we address the problems of fully automatic localization and segmentation of 3D vertebral bodies from CT/MR images. We propose a learning-based, unified random forest regression and classification framework to tackle these two problems. More specifically, in the first stage, the localization of 3D vertebral bodies is solved with random forest regression where we aggregate the votes from a set of randomly sampled image patches to get a probability map of the center of a target vertebral body in a given image. The resultant probability map is then further regularized by Hidden Markov Model (HMM) to eliminate potential ambiguity caused by the neighboring vertebral bodies. The output from the first stage allows us to define a region of interest (ROI) for the segmentation step, where we use random forest classification to estimate the likelihood of a voxel in the ROI being foreground or background. The estimated likelihood is combined with the prior probability, which is learned from a set of training data, to get the posterior probability of the voxel. The segmentation of the target vertebral body is then done by a binary thresholding of the estimated probability. We evaluated the present approach on two openly available datasets: 1) 3D T2-weighted spine MR images from 23 patients and 2) 3D spine CT images from 10 patients. Taking manual segmentation as the ground truth (each MR image contains at least 7 vertebral bodies from T11 to L5 and each CT image contains 5 vertebral bodies from L1 to L5), we evaluated the present approach with leave-one-out experiments. Specifically, for the T2-weighted MR images, we achieved for localization a mean error of 1.6 mm, and for segmentation a mean Dice metric of 88.7% and a mean surface distance of 1.5 mm, respectively. For the CT images we achieved for localization a mean error of 1.9 mm, and for segmentation a mean Dice metric of 91.0% and a mean surface distance of 0.9 mm, respectively.
《Automated medical image segmentation techniques》
Accurate segmentation of medical images is a key step in contouring during radiotherapy planning. Computed topography (CT) and Magnetic resonance (MR) imaging are the most widely used radiographic techniques in diagnosis, clinical studies and treatment planning. This review provides details of automated segmentation methods, specifically discussed in the context of CT and MR images. The motive is to discuss the problems encountered in segmentation of CT and MR images, and the relative merits and limitations of methods currently available for segmentation of medical images.
《基於改進高斯混合模型的MR圖像分割》
傳統高斯混合模型分割核磁共振圖像時嚴重依賴初值，且易受圖像中偏移場與噪聲的影響。為此，提出一種基於片信息的改進高斯混合模型。采用模糊C均值聚類方法優化初始值，以減小初值對分割結果的影響，加快算法的收斂速度。使用Legendre多項式對偏移場進行擬合，並融入EM框架中，得到光滑的偏移場。利用鄰域信息降低噪聲的影響，使模型在降低噪聲影響的同時，保留細長拓撲結構信息。實驗結果表明，該模型能恢復出偏移場，分割結果較好。
針對Level Set算法運算速度較慢和易產生邊緣泄露的不足，引入了結合快速步進的Level Set算法，提出了一套完整的分割人體足部骨骼圖像技術路線。修正了原始。光切片”圖像噪聲多的不足，通過預處理去除噪聲，增強邊緣；設定分割初始點和運算參數，運行改進的Level Set算法提取骨骼區域；運行形態學開操作進行邊緣斷裂和毛刺修復。實驗結果表明，該處理流程具有較好的准確度和魯棒性，與經典Level Set算法相比，改進的算法能提高19*／o--36％的運行速度．
《Fast, accurate, and fully automatic segmentation of the right ventricle in short-axis cardiac MRI 》
This paper presents a fully automatic method to segment the right ventricle (RV) from shortaxis cardiac MRI. A combination of a novel window-constrained accumulator thresholding technique,binary Difference of Gaussian (DoG) filters, optimal thresholding, and morphology are utilized to drive the segmentation. A priori segmentation window constraints are incorporated to guide and refine the process, as well as to ensure appropriate area confinement of the segmentation. Training and testing were
performed using a combined 48 patient datasets supplied by the organizers of the MICCAI 2012 Right Ventricle Segmentation Challenge, allowing for unbiased evaluations and benchmark comparisons.Marked improvements in speed and accuracy over the top existing methods are demonstrated.
《V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation》
Convolutional Neural Networks (CNNs) have been recently employed to solve problems from both the computer vision and medical image analysis fields. Despite their popularity, most approaches are only able to process 2D images while most medical data used in clinical practice consists of 3D volumes. In this work we propose an approach to 3D image segmentation based on a volumetric, fully convolutional, neural network. Our CNN is trained end-to-end on MRI volumes depicting prostate, and learns to predict segmentation for the whole volume at once. We introduce a novel objective function, that we optimise during training, based on Dice coefficient. In this way we can deal with situations where there is a strong imbalance between the number of foreground and background voxels. To cope with the limited number of annotated volumes available for training, we augment the data applying random non-linear transformations and histogram matching. We show in our experimental evaluation that our approach achieves good performances on challenging test data while requiring only a fraction of the processing time needed by other previous methods.