一种自动寻找染色体着丝粒的算法

染色体着丝粒的自动定位，是计算机自动染色体核型分析的关键。本文提出了一种染色体着丝粒自动寻找定位算法。该算法应用于实验用标准人体染色体图片，正确定位率达９０％。     

染色体核型分析深度学习方法综述

染色体核型分析是细胞遗传学领域重要的实验技术，并逐步在包括生殖医学在内的诸多现代临床领域和科学研究方面得到广泛应用，但即使是经验丰富的细胞遗传学家也需要大量时间才能完成染色体核型分析。基于传统方法的染色体核型自动化分析方法精度较低，仍需要细胞遗传学家花费大量时间、精力校正。目前基于深度学习的染色体核型自动分析方法成果较多，但缺乏对该领域现状的总结、对未来发展的展望等。因此，本文对基于深度学习的染色体核型自动分析方法进行综述，归纳总结了现有的研究分析任务，挑选了具有代表性的方法并梳理解决方案，展望了未来发展方向。通过整理发现，基于深度学习的染色体核型自动化分析方法取得了很多成果，但仍存在一些问题。首先，已有的中文综述性工作仅集中于某一子领域或者调研不够全面和深入。其次，染色体核型分析任务与临床紧密结合，受各种因素制约，任务类型繁多，解决方案复杂，难以窥斑见豹。最后，现有方法主要集中于染色体分类和染色体分割任务，而诸如染色体计数、染色体预处理等任务研究成果较少，需要厘清问题，吸引更多研究人员关注。综上所述，基于深度学习的染色体核型自动分析方法仍有较大发展空间。     

舌体歪斜的自动分析

舌形态的客观分析是中医客观化、现代化的组成部分。舌体歪斜与肌体脏腑器官病理程度相关。该文提出一种新的确定对称中轴线的方法,并结合嘴角定位,实现了舌体歪斜的自动定量分析。实验结果表明了该方法的有效性。     

染色体自动分割算法

染色体的自动分割是染色体自动分析系统中最重要的步骤.因此对典型的染色体自动分割算法进行了研究,分析了算法的原理及实现,对算法的结果进行比较和讨论,并分析其中存在的问题.结果表明,基于几何分析的自动分割算法在粘连染色体的分割方面具有良好的效果.     

人类染色体识别系统

文章的研究应用于清华大学电子系开发的CIARS犤9犦系统。文章介绍了基于自动分割的染色体识别系统的流程。对于人类染色体的自动分割一直是染色体识别的一个难题。由于染色体的柔软性和样本的离散性,染色体的样本中往往存在着各种交叠和粘连现象。目前市场上的各种染色体处理系统均未提供染色体的完全自动分割,而采用了人机交互的方式。由于染色体不是刚体,普通的对于刚体的分割方法无法应用于染色体交叠和粘连的分割。文章介绍一种基于边界分析的分割方法,成功地将分割问题转化为模式识别问题。     

支持动态编辑的交通路网语义建模方法

针对现有交通路网语义建模方法不支持新增道路、道路加宽/变窄、潮汐车道设置、 道路连通性设置等路网交通属性动态编辑修改的问题,提出了一种支持动态编辑的交通路网语 义建模方法。通过分析路网内正常路段和交叉路口之间的耦合关系,给出了由传统 Lane、路口 Lane,Link,Connection,Intersection 以及 Road 等组成的具有层次耦合关系的路网语义数据。 只需输入道路中轴线数据,上述语义数据即可全部自动生成。路网交通属性动态编辑修改后, 只需对相应路段和相连路口语义进行解耦重新计算即可完成路网语义数据的动态更新。实验结 果表明,只需要输入道路中轴线矢量数据和道路宽度数据,该方法即可准确、高效地完成交通 路网语义建模,并且支持路网交通属性的动态编辑且可以实现交通相位的自动生成。     

人类显带染色体的图象分析与识别系统

本文介绍了我们自行研究开发的人类显带染色体的图象分析与识别系统的原理、结构和性能，概述了系统使用的核心技术和算法，包括显带染色体的自动分割算法、显带染色体的特征提取算法和分类器的设计方法。     

基于纹理映射的动态实时虚拟现实方法

提出了一种动态实时的进行纹理映射的方法,该方法首先通过自动寻找三维旋转模型主中轴线来与纹理图片的中垂线进行配准,然后沿左右方向进行纹理保长等单位映射.该方法可将纹理图片等单位的映射到三维模型上,同时可以进行模型修正和纹理图片的动态移动,纹理不会变形,这完全不同于其它指定多边形及相应纹理坐标进行插值的纹理映射方法,后者将无法与用户进行交互且无法动态贴图.实现结果表明,该方法很好地模拟了现实世界的贴图效果,且具有很好的灵活性、实用性和通用性,可以应用到陶瓷贴花的虚拟现实软件中,进行提前演示效果.     

基于机器视觉的维氏硬度自动测量方法

提出一种自动分析维氏硬度压痕的显微放大图像的方法来测量硬度,克服了以往方法中将压痕边缘当做直线来拟合所带来的测量误差。压痕边缘有一定程度弯曲,用直线来拟合边缘会引入较大误差。考虑到压痕边缘在顶点附近的区域内高度近似为直线,提出了先寻找顶点所在区域,再逐步缩小寻找范围在各区域内寻找其精确位置的方法。在VC＋＋平台上实现了对维氏硬度压痕图进行自动分析和自动计算。实验表明,该方法的测量结果与标称值十分接近,测量重复性好。     

基于数学形态学的染色体自动计数算法

通过研究染色体自动分析技术在评估儿童急性淋巴细胞白血病预后中的应用,提出了一种基于数学形态学的染色体自动计数算法.算法首先对输入的染色体图像进行了预处理,接着用标号法给图像中的每个对象赋予惟一的标号.为了消除图像中噪声对统计结果的影响,按照设定的规则将噪声对象删除.对于含有粘连染色体的对象,通过腐蚀运算计算出实际染色体个数,并根据结果修正标号值,修正结果即为图像中的杂色体数目.实验证明,该算法可以自动统计出图像中的染色体数目.     

The algorithm of the cockshafer walk: a movement automaton applied to the construction of the integrated optical density profile of prometaphase chromosomes

We present an algorithm designed to obtain the optical density profile of prometaphase chromosomes. Our movement automation (the cockshafer) is able to move along the median axis of the chromosome. The criterion used to maintain the correct movement direction lies on the distances from the automaton position to the side boundaries (lengths of the cockshafer antennae). The automaton movements are regulated by rules defined in a decision repertory. This new method does not use a transformation of the image and does not require any prior knowledge of the shape of the chromosome to be sampled.     

基于三维交叉阵的相干分布式信号源DOA估计

研究了相干分布式信号源的二维中心波达方向估计（Direction of arrival, DOA）。利用三维交叉阵的对称特性,用求根的方法和广义ESPRIT算法分别估计出相干分布式信号源的中心俯仰角和中心方位角。所提算法无需知道相干分布式信号源的角信号分布函数,并且只需要一维谱搜索。此外,非圆信号的引入使得算法获得了更高的估计精度。计算机仿真验证了算法的有效性。     

Quantum novel genetic algorithm based on parallel subpopulation computing and its application

A quantum novel genetic algorithm based on subpopulation parallel computing is presented, where quantum coding and rotation angle are improved to inspire more efficient genetic computing methods. In the algorithm, each axis of the solution space is divided into k parts, the individual (or chromosome) from each different subspace being coded differently, and the probability amplitude of each quantum bit or Q-bit is regarded as two paratactic genes. The basic quantum computing theory and classical quantum genetic algorithm are briefly introduced before a novel algorithm is presented for the function optimum and PID problem. Through a comparison between the novel algorithm and the classical counterpart, it is shown that the quantum inspired genetic algorithm performs better on running speed and optimization capability.     

A modular framework for the automatic classification of chromosomes in Q-band images

The manual analysis of the karyogram is a complex and time-consuming operation, as it requires meticulous attention to details and well-trained personnel. Routine Q-band laboratory images show chromosomes that are randomly rotated, blurred or corrupted by overlapping and dye stains. We address here the problem of robust automatic classification, which is still an open issue. The proposed method starts with an improved estimation of the chromosome medial axis, along which an established set of features is then extracted. The following novel polarization stage estimates the chromosome orientation and makes this feature set independent on the reading direction along the axis. Feature rescaling and normalizing techniques take full advantage of the results of the polarization step, reducing the intra-class and increasing the inter-class variances. After a standard neural network based classification, a novel class reassignment algorithm is employed to maximize the probability of correct classification, by exploiting the constrained composition of the human karyotype. An average 94% of correct classification was achieved by the proposed method on 5474 chromosomes, whose images were acquired during laboratory routine and comprise karyotypes belonging to slightly different prometaphase stages. In order to provide the scientific community with a public dataset, all the data we used are publicly available for download.     

Chromosome classification based on the band profile similarity along approximate medial axis

Automated chromosome classification is an essential task in cytogenetics and has been an important pattern recognition problem. Numerous attempts were made in the past to characterize chromosomes for the purposes of clinical and cancer cytogenetics research. It is important to determine good features and develop feature extraction schemes for chromosome classification. In this paper we propose efficient approaches for medial axis determination and profile matching of human chromosomes without identifying centromeres. The medial axis determination is based on simple cross-section analysis. The features of the band profile obtained along the axis are then used to classify a chromosome based on a subsequence matching technique. Using a special indexing structure, we are able to perform fast similarity search and dynamic insertion and deletion over the established subsequence database of chromosome profiles. According to the experimental results, the developed adaptive system can automatically and efficiently determine the medial axis of a given chromosomes, and achieve satisfactory classification results.     

Computation of medial axis and offset curves of curved boundaries in planar domain

In this paper, we begin our research from the generating theory of the medial axis. The normal equidistant mapping relationships between two boundaries and its medial axis have been proposed based on the moving Frenet frames and Cesaro’s approach of the differential geometry. Two pairs of adjoint curves have been formed and the geometrical model of the medial axis transform of the planar domains with curved boundaries has been established. The relations of position mapping, scale transform and differential invariants between the curved boundaries and the medial axis have been investigated. Based on this model, a tracing algorithm for the computation of the medial axis has been generated. In order to get the accurate medial axis and branch points, a Two_Tangent_Points_Circle algorithm and a Three_Tangent_Points_Circle algorithm have been generated, which use the results of the tracing algorithm as the initial values to make the iterative process effective. These algorithms can be used for the computation of the medial axis effectively and accurately. Based on the medial axis transform and the envelope theory, the trimmed offset curves of curved boundaries have been investigated. Several numerical examples are given at the end of the paper.     

Mandelbrot集高周期混沌吸引子定位算法研究

研究Mandelbrot集混沌分形图谱混沌吸引子定位算法.设计高周期混沌吸引子的定位算法,该算法将复平面区域网格化,根据混沌吸引子模值局部最小的特性,确定其坐标位置.针对实轴上混沌吸引子分布密集难以处理的情况,算法将复平面划分为实轴上侧、实轴和实轴下侧三个部分,实轴上的混沌吸引子采用单独的遍历方法进行查找.分析定位算法...     

基于分层的二视图乳腺病灶区域匹配研究

乳腺图像的非刚体特性使其不能通过刚体的匹配方法进行匹配。为此,提出一种基于分层算法的肿块匹配方法。利用最大类间方差法进行阈值分割以截取胸肌区域,使用最小二乘法进行胸壁线的拟合,找出乳头以及中轴线的位置,建立局部坐标系确定匹配条形区域带。采用分层算法在条形区域带内进行疑似病灶区域的提取,通过加权互信息相似性度量实现肿块匹配。对分层算法肿块匹配方法进行可行性分析和实验验证。选取100对图像进行肿块匹配实验,结果表明,相对于非分层算法,分层算法的匹配结果更有效,匹配精度达到86%。     

Computerized axial tomographic reconstruction of coronary tree cross sections from a small number of cineradiographic views

Three-dimensional reconstruction of the coronary tree from a conventional cineangiographic study is limited, in part, by the small number of available cineangiographic views. The potential utility of a maximum entropy iterative algorithm (MENT) for reconstruction of myocardial planes perpendicular to the axis of cineangiographic rotation from a small number (n = 6-18) of cineangiographic views was tested in vitro. The coronary arteries of postmortem human, canine, and calf hearts were filled with a silicone/thorium oxide mold to simulate in vivo angiographic contrast. Thirty-five-millimeter cineradiographs of each heart were obtained at 10 degrees intervals over 180 degrees about a myocardial central axis of rotation under exposure conditions which simulated the clinical setting. Projection data were derived from cinedensitometric scans across the entire myocardial shadow, perpendicular to the axis of rotation in each view, after 512 X 512 digitization with a vidicon camera/digitizer interfaced to a VAX computer. Comparison of MENT-reconstructed images with corresponding anatomic myocardial cross sections indicate that as few as 6 to 12 views can be used to reconstruct the cross sections of the multiple coronary branches (n = 6-11) within a plane of reconstruction.     

求解广义旅行商问题的混合染色体遗传算法

提出了针对广义旅行商问题(GTSP)的混合染色体遗传算法(HCGA)。目前,广义染色体遗传算法(GCGA)是求解GTSP问题的最好方法,但这种方法在编码设计上存在不足,使得算法全局搜索能力较差。在GCGA算法基础上,设计了二进制和整数混合编码的染色体,并更新了交叉和变异算子设计,得到改进算法HCGA。理论和实验结果都说明:HCGA比GCGA等多种算法具有更强的全局搜索能力。