醫(yī)學(xué)圖像配準(zhǔn)的關(guān)鍵技術(shù)研究
發(fā)布時(shí)間：2018-08-12 08:59
【摘要】：隨著醫(yī)學(xué)成像技術(shù)的發(fā)展,越來(lái)越多的臨床應(yīng)用要求對(duì)來(lái)自不同主體,不同時(shí)期或不同成像設(shè)備的醫(yī)學(xué)圖像進(jìn)行比較和分析。醫(yī)學(xué)圖像配準(zhǔn)是醫(yī)學(xué)圖像分析和計(jì)算解剖的一個(gè)關(guān)鍵步驟,被廣泛應(yīng)用在疾病診斷、手術(shù)導(dǎo)航、人腦圖譜和各種醫(yī)學(xué)評(píng)價(jià)等方面。然而,醫(yī)學(xué)圖像的多樣性、復(fù)雜性和非連續(xù)性等特征使得醫(yī)學(xué)圖像配準(zhǔn)技術(shù)具有很大的挑戰(zhàn)性。醫(yī)學(xué)圖像數(shù)據(jù)(如擴(kuò)散張量圖像)通常都是非線性結(jié)構(gòu),存在于非線性的流形上。已有的配準(zhǔn)技術(shù),無(wú)論單模態(tài)還是多模態(tài),剛體還是非剛體,基于參數(shù)還是非參數(shù),要么忽略流形的非線性幾何,直接在線性的歐式空間下進(jìn)行研究,要么對(duì)非線性數(shù)據(jù)結(jié)構(gòu)所包含的豐富的空間信息考慮不夠。然而這些信息對(duì)空間變換下圖像拓?fù)浣Y(jié)構(gòu)的保持具有重要意義。本文較系統(tǒng)地對(duì)幾種醫(yī)學(xué)成像技術(shù)進(jìn)行了研究,尤其是磁共振成像技術(shù)和擴(kuò)散張量成像技術(shù)。以拓?fù)鋵W(xué)、微分幾何和幾何代數(shù)作為空間分析的數(shù)學(xué)工具,對(duì)醫(yī)學(xué)數(shù)據(jù)的特征拓?fù)浣Y(jié)構(gòu)以及數(shù)據(jù)間的空間關(guān)系進(jìn)行深入探討,圍繞配準(zhǔn)算法的魯棒性,精度和拓?fù)浔３中詫?duì)非線性醫(yī)學(xué)成像數(shù)據(jù)的高維空間分布關(guān)系進(jìn)行研究。本文的主要貢獻(xiàn)如下。(1)傳統(tǒng)的非參數(shù)微分同胚配準(zhǔn)算法只是基于像素灰度恒定的假設(shè),忽略了高維空間變換中數(shù)據(jù)流形的非線性結(jié)構(gòu)的豐富性和拓?fù)湫詫?duì)保持合理物理結(jié)構(gòu)的影響。本文在微分同胚Demons算法的基礎(chǔ)上,提出了一種局部自適應(yīng)拓?fù)浔３值腗R圖像配準(zhǔn)。為了獲得更豐富的空間信息和幾何結(jié)構(gòu),首先構(gòu)造正定對(duì)稱矩陣,并在一定條件下形成高維非線性的李群流形,然后利用流形學(xué)習(xí)方法進(jìn)行自適應(yīng)的鄰域選擇,從而更精確地逼近流形的線性切空間,保持流形的非線性結(jié)構(gòu),使圖像特征空間的拓?fù)浣Y(jié)構(gòu)在非線性的微分同胚變換中更好地保持物理合理性。(2)針對(duì)傳統(tǒng)的DT圖像中張量的重定向策略只適合于剛體配準(zhǔn)或因迭代產(chǎn)生的計(jì)算代價(jià),本文把張量集轉(zhuǎn)換成一種點(diǎn)集的規(guī)范式,提出一種規(guī)范式下的DT圖像仿射配準(zhǔn)。在這種規(guī)范式下,仿射變換下兩個(gè)張量集之間的配準(zhǔn)就可以轉(zhuǎn)換為旋轉(zhuǎn)變換下規(guī)范式之間的配準(zhǔn),但仍然保持了仿射變換中的非剛體形變分量對(duì)重定向的影響,使得形變?cè)诮馄式Y(jié)構(gòu)上更合理。傳統(tǒng)的基于重定向的DT圖像仿射配準(zhǔn)算法只提取剛體旋轉(zhuǎn)部分進(jìn)行空間變化,忽略平移,縮放和切變等形變分量對(duì)仿射變換的影響。所以比較傳統(tǒng)方法,本文算法用于仿射變換能獲得更好地精度。進(jìn)一步,為了改善由于重定向引起的計(jì)算代價(jià),使優(yōu)化過(guò)程更有效率,利用旋轉(zhuǎn)變換群---李群SO(3)描述彌散張量的特殊數(shù)學(xué)結(jié)構(gòu),用單位四元數(shù)旋轉(zhuǎn)替代三維旋轉(zhuǎn)矩陣,從絕對(duì)定向中找到最優(yōu)閉式解,能大大減少計(jì)算代價(jià)。(3)針對(duì)傳統(tǒng)多模態(tài)配準(zhǔn)方法忽視圖像的空間結(jié)構(gòu)和像素間的空間關(guān)系和假定灰度全局一致,提出一種基于學(xué)習(xí)理論的多模態(tài)圖像配準(zhǔn)算法。本文利用自回歸線性動(dòng)態(tài)模型描述圖像的局部高維非線性空間結(jié)構(gòu),這使得特征空間包含了更多空間信息。然后通過(guò)參數(shù)化動(dòng)態(tài)模型構(gòu)造出具有李群結(jié)構(gòu)的群元素,并形成黎曼流形,接下來(lái)把黎曼流形嵌入到更高維的再生核希爾伯特空間,再在核空間引入核函數(shù)尋找最優(yōu)的相似性測(cè)度,這種核技巧可以把非線性數(shù)據(jù)映射到一個(gè)隱式的高維中進(jìn)行處理。算法不僅對(duì)剛體配準(zhǔn),而且對(duì)仿射配準(zhǔn)也適用。
[Abstract]:With the development of medical imaging technology, more and more clinical applications require the comparison and analysis of medical images from different subjects, different periods or different imaging equipments. However, the diversity, complexity and discontinuity of medical images make medical image registration challenging. Medical image data (such as diffusion tensor images) are usually nonlinear structures and exist on nonlinear manifolds. Rigid or non-rigid, based on parameters or non-parameters, or ignoring the nonlinear geometry of manifolds, directly study in linear Euclidean space, or the rich spatial information contained in the non-linear data structure is not considered enough. Several medical imaging techniques, especially magnetic resonance imaging and diffusion tensor imaging, are systematically studied. Topology, differential geometry and geometric algebra are used as mathematical tools for spatial analysis. The characteristic topological structure of medical data and the spatial relationship between medical data are discussed in detail. The robustness of registration algorithm is focused on. The main contributions of this paper are as follows. (1) The traditional non-parametric differential homeomorphism registration algorithm is based on the assumption that the pixel gray level is constant, ignoring the richness and topological pairs of the nonlinear structure of the data manifold in the high-dimensional space transformation. Based on the differential homeomorphism Demons algorithm, a locally adaptive topology preserving MR image registration algorithm is proposed in this paper. In order to obtain more spatial information and geometric structure, a positive definite symmetric matrix is constructed, and a high-dimensional nonlinear Lie group manifold is formed under certain conditions, and then a manifold is used. Adaptive neighborhood selection is used to approximate the linear tangent space of the manifold more accurately and keep the nonlinear structure of the manifold, so that the topological structure of the image feature space can keep the physical rationality better in the nonlinear differential homeomorphism transformation. (2) The traditional DT image tensor redirection strategy is only suitable for rigid bodies. In this paper, tensor set is transformed into a normal form of point set, and a new affine registration method for DT images is proposed. In this form, the registration between two tensor sets under affine transformation can be transformed into the registration between the normalizations under rotational transformation, but the affine transformation is still preserved. The influence of non-rigid deformation components on redirection makes the deformation more reasonable in anatomical structure. The traditional affine registration algorithm based on redirection only extracts the rotational part of rigid body for spatial change, ignoring the effects of translation, scaling and shear on affine transformation. Further, in order to improve the computational cost caused by redirection and make the optimization process more efficient, the special mathematical structure of the dispersion tensor is described by the rotational transformation group-Lie group SO(3). The unit quaternion rotation is used instead of the three-dimensional rotation matrix to find the optimal closed-form solution from the absolute orientation, which can be greatly reduced. (3) Aiming at the neglect of spatial structure and spatial relationship between pixels and the assumption that gray level is globally consistent in traditional multi-modal registration methods, a multi-modal image registration algorithm based on learning theory is proposed. Spaces contain more spatial information. Then, group elements with Lie group structure are constructed by parameterized dynamic model, and Riemannian manifolds are formed. Next, Riemannian manifolds are embedded into higher dimensional reproducing kernel Hilbert space, and kernel functions are introduced into kernel space to find the optimal similarity measure. The algorithm is not only suitable for rigid registration, but also for affine registration.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41
本文編號(hào)：2178575
[Abstract]:With the development of medical imaging technology, more and more clinical applications require the comparison and analysis of medical images from different subjects, different periods or different imaging equipments. However, the diversity, complexity and discontinuity of medical images make medical image registration challenging. Medical image data (such as diffusion tensor images) are usually nonlinear structures and exist on nonlinear manifolds. Rigid or non-rigid, based on parameters or non-parameters, or ignoring the nonlinear geometry of manifolds, directly study in linear Euclidean space, or the rich spatial information contained in the non-linear data structure is not considered enough. Several medical imaging techniques, especially magnetic resonance imaging and diffusion tensor imaging, are systematically studied. Topology, differential geometry and geometric algebra are used as mathematical tools for spatial analysis. The characteristic topological structure of medical data and the spatial relationship between medical data are discussed in detail. The robustness of registration algorithm is focused on. The main contributions of this paper are as follows. (1) The traditional non-parametric differential homeomorphism registration algorithm is based on the assumption that the pixel gray level is constant, ignoring the richness and topological pairs of the nonlinear structure of the data manifold in the high-dimensional space transformation. Based on the differential homeomorphism Demons algorithm, a locally adaptive topology preserving MR image registration algorithm is proposed in this paper. In order to obtain more spatial information and geometric structure, a positive definite symmetric matrix is constructed, and a high-dimensional nonlinear Lie group manifold is formed under certain conditions, and then a manifold is used. Adaptive neighborhood selection is used to approximate the linear tangent space of the manifold more accurately and keep the nonlinear structure of the manifold, so that the topological structure of the image feature space can keep the physical rationality better in the nonlinear differential homeomorphism transformation. (2) The traditional DT image tensor redirection strategy is only suitable for rigid bodies. In this paper, tensor set is transformed into a normal form of point set, and a new affine registration method for DT images is proposed. In this form, the registration between two tensor sets under affine transformation can be transformed into the registration between the normalizations under rotational transformation, but the affine transformation is still preserved. The influence of non-rigid deformation components on redirection makes the deformation more reasonable in anatomical structure. The traditional affine registration algorithm based on redirection only extracts the rotational part of rigid body for spatial change, ignoring the effects of translation, scaling and shear on affine transformation. Further, in order to improve the computational cost caused by redirection and make the optimization process more efficient, the special mathematical structure of the dispersion tensor is described by the rotational transformation group-Lie group SO(3). The unit quaternion rotation is used instead of the three-dimensional rotation matrix to find the optimal closed-form solution from the absolute orientation, which can be greatly reduced. (3) Aiming at the neglect of spatial structure and spatial relationship between pixels and the assumption that gray level is globally consistent in traditional multi-modal registration methods, a multi-modal image registration algorithm based on learning theory is proposed. Spaces contain more spatial information. Then, group elements with Lie group structure are constructed by parameterized dynamic model, and Riemannian manifolds are formed. Next, Riemannian manifolds are embedded into higher dimensional reproducing kernel Hilbert space, and kernel functions are introduced into kernel space to find the optimal similarity measure. The algorithm is not only suitable for rigid registration, but also for affine registration.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41
【參考文獻(xiàn)】
相關(guān)期刊論文 前2條
1 閆德勤;劉彩鳳;劉勝藍(lán);劉德山;;大形變微分同胚圖像配準(zhǔn)快速算法[J];自動(dòng)化學(xué)報(bào);2015年08期
2 許鴻奎;江銘炎;楊明強(qiáng);;基于改進(jìn)光流場(chǎng)模型的腦部多模醫(yī)學(xué)圖像配準(zhǔn)[J];電子學(xué)報(bào);2012年03期
相關(guān)博士學(xué)位論文 前1條
1 周志勇;醫(yī)學(xué)圖像非剛性配準(zhǔn)方法研究[D];中國(guó)科學(xué)院研究生院（長(zhǎng)春光學(xué)精密機(jī)械與物理研究所）;2013年
，本文編號(hào)：2178575
本文鏈接：http://sikaile.net/kejilunwen/ruanjiangongchenglunwen/2178575.html
最近更新
教材專著
