结合时域信息的区域生长算法及其在动脉超声造影图像分割中的应用

为了有效地提取动脉超声造影图像中的血管边界,提高动脉内膜附近血流流场超声测量的准确性,提出一种结合时域信息的区域增长算法进行动脉超声造影图像的血管边界分割.该算法根据超声造影微泡在血液内的流动特性,利用连续两帧图像之间的灰度差异构造窗帧差,令其作为生长条件,结合邻域灰度进行区域生长;采用数学形态学的闭合运算与一种自定义的边界平滑方法对区域生长结果进行处理,填补了区域生长中产生的空洞并平滑了不规则边界.最后通过大鼠颈动脉的超声造影图像实验,验证了文中算法的有效性和可行性,该算法能够准确地提取出模糊的动脉边界,时间复杂度低.     

微型机超声图像处理系统的设计

本文介绍了对超声图像的获取、动态显示、伪彩色处理、存贮、回放等方法及系统管理功能的设计。     

超声序列图像实时体渲染系统设计与实现

设计并实现了对超声序列图像实时体渲染的可视化系统。该系统能够对目前的二维超声设备进行扩展,利用对视频数据的重采样和预处理得到三维切片体数据,并通过GPU加速达到了对DDR与MINIP光照模型的实时体渲染。阐述系统的设计模块与基于VTK的实现过程,最后通过对实际肝脏超声数据实验结果说明了系统对二维超声数据分析的有效性。     

改进Snake模型在医学超声图像分割中的应用

因其本身所具有的复杂性,医学超声图像的分割足一件非常困难的事情.Snake算法能够有效地利用图像局部与整体的信息,实现对边界的准确定位,因此非常适用于医学超声图像的处理.针对传统Snake模型对初始轮廓依赖性的问题,根据图像边缘附近灰度差异显著的特性,提出了获取初始轮廓点的方法--扇形法,运用方法可以获得靠近日标边界的初始轮廓点,并且得到的轮廓点是有序点集.能量最小化过程运用贪婪算法来获得图像的特征边缘点,最后采用分段多项式拟合的方法来获得连续的图像边缘.实验结果表明.所采用的方法可以获得连续、封闭的边缘曲线,能够较好的将目标从图像中提取出来.     

多阶U-Net甲状腺超声图像自动分割方法

甲状腺超声图像广泛应用于甲状腺相关疾病的诊断。针对甲状腺超声图像对比度低、边缘模糊以及散斑噪声严重等问题,提出一种基于多阶U-Net的深度卷积网络模型,用于实现甲状腺腺体与甲状腺结节的自动分割。该模型以U-Net为基本网络框架,通过不断进阶的特征融合,以实现图像边缘的信息提取。同时,在模型中使用了一种多尺度残差卷积模块以进一步提升分割精度。对比实验结果表明,该模型相较于其他方法能够获得更好的分割结果,具有一定的临床应用价值。     

基于随机游走的医学超声图像分割

医学超声图像不可避免地存在斑点噪声、弱边界等问题,很难达到满意的分割效果。随机游走算法对噪声具有鲁棒性,对弱边界有良好的提取能力。将此算法应用于医学超声图像分割,通过融合区域信息与用户指定的种子点信息,借助于电路模拟以及组合Dirichlet问题,可以得到每个非种子点到标记了目标点或者背景点的概率,并对其赋予概率中最大的种子点所对应的标记,从而实现图像的分割。实验结果表明,该方法对医学超声图像的分割是有效的。此算法通过求解稀疏的、对称的、正定的线性方程的系统来获得Dirichlet问题的解,使计算速度大为提高。     

双态形状重构及其在前列腺超声图像分割中的应用

前列腺超声图像在临床中的准确分割对于后续诊断具有重要的影响,而当前已有研究结论无法精确分割各个部分。提出了一种基于点分布模型和流形学习的双态形状重构的方法,并对前列腺超声图像进行分割：通过随机森林指示隐态表达进行目标初定位;改进边界算子以改善粗分割准确性;使用显态表达与噪声部分相邻的部分灰度显著点来进行插值计算,从而恢复整体形状。该分割方式不仅减少了数据计算量,还增加了分割可靠性。实验表明,该方法的DSC指标为97.38%,mIoU指标为95.24%,精度强于当前热门分割神经网络。     

基于边缘检测的图像分割的超声诊断机器人控制系统设计

传统系统在嵌入信息后峰值信噪比较高,使图像分割结果与实际效果相差较大,导致控制系统超声诊断结果并不精准,针对该问题,提出基于边缘检测的图像分割的超声诊断机器人控制系统设计。在控制系统的硬件结构部分,以STM32F103C8T6单片机为核心的控制器,使用型号为3MC58的步进电机驱动器,通过角位移量控制脉冲个数,以实现更加准确的定位。在软件设计中,通过图像分割技术检测图像边缘特征,提高图像分割精度和准确度,帮助诊断机器人提高诊断功能。分析不同嵌入率下峰值信噪比,在确定该值后将传统系统与该系统的超声诊断功能对比分析,由实验结果可知,该系统在嵌入信息后峰值信噪比较高的情况下也能保证图像分割结果与实际效果一致,对0.8 bpp嵌入率图像的分割处理效果平均值为89%,超声诊断准确率平均值为88.6%,表明该系统的控制性能较好,能够为医学、航天航空及军事领域提供设备支持。     

基于弹性匹配活动轮廓模型心脏超声图像分割

超声医学图像由于受成像机理的影响,图像对比度不高、边缘不明显.基于传统活动轮廓线模型(snake模型)的分割方法可能产生过分割或泄漏问题.由于医学图像中拓扑结构已知,因此基于先验知识的活动轮廓线分割方法是解决这个问题的一个有效途径.建立一种新的基于弹性匹配活动轮廓线模型,该方法将待分割曲线的形状与原型曲线用弹性匹配测量变形量或相似度.曲线在演化过程中,根据变形量或相似度,可以准确分割模糊的边缘,同时保持整体目标分割形状.通过对二维小儿超声心脏图像的左心房内壁进行分割,经比较,基于弹性匹配活动轮廓线分割比传统活动轮廓线分割的误差有显著减少,避免了传统活动轮廓线的过分割或泄漏问题.     

基于图形硬件加速的Freehand三维超声图像滤波与插值技术

本文对采用Freehand超声采集方法得到体数据进行三维重建与可视化问题进行了研究。由于传统基于CPU的方法在数据预处理与三维成像的过程中要耗费大量时间,本文应用可编程图形硬件(GPU)技术,实现快速的纹理查找与计算,并进行图像滤波与插值运算。由于可编程图形硬件将很多算法用硬件并行计算,可以明显缩短成像时间。最后比较了两种滤波方法和一种插值算法来提高三维图像的质量,实验证明,基于GPU的同中心插值法能减少噪声信号干扰,缩短绘制时间。     

The Improvement of Segmentation of Lung Pathologies and Pleural Effusion on CT-scans of Patients with Covid-19

Programming and Computer Software - In 2020 the outbreak of Covid-19 influenced lives of billions of people all around the globe and motivated governments of different countries to revisit the...     

Introduction to proceedings of the workshop on Knudsen Effusion Mass Spectrometry

The collected papers in this Special Edition are from a Workshop on one of the most informative methods of experimental thermodynamics, “Knudsen Effusion Mass Spectrometry” (KEMS). This international workshop was held in Jülich, Germany in October of 2017. KEMS is based on the near equilibrium conditions between a condensed phase and a vapor that is achieved in a Knudsen cell. The well-defined orifice on the Knudsen cell forms a molecular beam, which is directed to a mass spectrometer. The measured ion intensities from each vapor species can be converted to partial pressures. These partial pressures give a full picture of the vapor composition and thermochemistry of the formed vapor species. For a multi-component sample, thermodynamics of the condensed phase can be attained as well. The measured partial pressures of each species allow a wealth of thermodynamic information to be collected: enthalpy, entropy, Gibbs energy, Helmholtz energy, activity of component i, equilibrium constants, and others. The widely used thermochemical tables contain many data entries from KEMS measurements. KEMS measurements are also used extensively in Calphad assessments.The 2017 KEMS workshop cover a wide range of applications for KEMS—thermodynamics of alloys, ceramics, and organic compounds for nuclear, chemical process, energy storage, aerospace propulsion, and geological fields. The wide versatility of KEMS was demonstrated—from low temperature alloys and organic compounds (700 K) to very high temperature oxides (2500 K). Vaporization and solution properties of both organic and inorganic compounds (halides, oxides, hydroxides, nitrides, carbides, etc.) were discussed. Papers on instrumentation development and improved sampling were also presented. There were discussions of new KEMS instruments and facilities under construction. It was shown that KEMS in conjunction with modeling is a valuable tool in thermodynamics. KEMS measurements have been used as an input to Calphad models and KEMS measurements have been used to guide and check first principles calculations.     

Proteomics analysis of serous fluids and effusions: Pleural, pericardial, and peritoneal

Serous body fluids are normally present in small amounts in the cavities lined by serous membranes, and effusions are the term usually used to represent an excess of these fluids. Thus, any changes in the effusions may accurately reflect various pathological processes. In this review, we will introduce the present status of effusion analysis. Proteomics allows the global study of protein expression and offers a comprehensive view to analyze effusion proteins at a system level. It deals with the study of proteins, their structures, localizations, PTM, functions, and interactions with other proteins. This paper reviews the variety of analytical methods that have been used for the analysis of effusions. The proteins expressed in effusions, which have been identified by proteomics methods are discussed.     

Quantification of thermodynamic properties for vaporisation reactions above solid Ga2O3 and In2O3 by Knudsen Effusion Mass Spectrometry

Even though Ga₂O₃ and In₂O₃ are broadly used as semi-conductors, thermodynamic data for their vaporisation reactions exhibit a large spread. Therefore, the vaporisation behaviour of solid Ga₂O₃ and In₂O₃ was determined by means of Knudsen Effusion Mass Spectrometry (KEMS). Ga₂O₃ and In₂O₃ were studied in an iridium Knudsen cell and heated over a temperature range of 1200–1750 K in order to identify the species present in the vapour phase, and determine their partial pressures. We find that M₂O (where M = Ga or In) is the most abundant gas species above the solid oxide, followed by M and MO, in accord with tabulated data. Following the calculation of partial pressures and equilibrium constants, we propose ${\Delta }_f{H}_{\mathrm{298,}3\mathrm{rd}}^{\mathrm{o}}\left(\mathrm{G}{\mathrm{a}}_2\mathrm{O}\left(\mathrm{g}\right)\right)$ = −68966 ± 7442 Jmol^-1 and ${\Delta }_f{H}_{\mathrm{298,}3\mathrm{rd}}^{\mathrm{o}}\left(\mathrm{I}{\mathrm{n}}_2\mathrm{O}\left(\mathrm{g}\right)\right)$ = −22245 ± 964 Jmol^-1 from the 3^rd law method. Deviations in ${\Delta }_f{H}_{\mathrm{298,}3\mathrm{rd}}^{\mathrm{o}}\left(i\right)$ relative to literature KEMS measurements are generally within ∼2% relative, and can be ascribed to the use of different ionisation cross sections, Knudsen cell material, temperature calibrations, as well as tabulated Gibbs energy functions. However, comparison with ab initio studies suggests the data reported in this work is more accurate than in previous studies, given that the ${\Delta }_f{H}_{\mathrm{298,}3\mathrm{rd}}^{\mathrm{o}}\left(\mathrm{I}\mathrm{n}\mathrm{O}\left(\mathrm{g}\right)\right)$ = 157744 ± 3681 Jmol^-1 deviates by only ∼0.2% from the theoretical value.     

变电站一次设备喷涂PRTV用量的分析

对国网宁夏电力检修公司宁安运维站近2年来一次设备喷涂PRTV用量的统计数据进行分析整理,得出各类一次设备按照标准要求喷涂PRTV时的用量估算值,并针对现存作业中存在的问题,提出相应解决方案。     

新型煤泥浮选机液位传感装置的研制

介绍了一种性能良好的新型浮球式煤浆液位传感装置。该装置利用浮球随着煤浆液位的变化而带动连杆作上下运动,通过机械装置将液位的变化转变成角位移传感器的轴的角度变化,分析角位移传感器输出信号可得到煤泥液位信息。经测试该传感装置各项指标达到了设计要求。该新型煤泥浮选机液位传感装置具有良好的动、静态特性,可以在选煤厂推广应用。     

制造型企业仓储物流管理的RFID应用——卷烟厂按垛出库及到货扫描系统初探

针对国家烟草管理的现状,有关主管部门在全国推行“行业卷烟生产经营决策管理系统工程”,利用条码等自动识别技术手段实时掌握全国的生产经营信息。但某卷烟厂此前的物流环节已经是“件烟成垛”运输,如何在尽可能保持原有企业管理体系的前提下,达到有关部门的数据统计要求,解决成垛卷烟的物流和信息流的交互与统一问题成为技改的核心。该项目成功的将条码识别与射频识别有机结合起来,为烟草行业信息化提供了生动的应用案例。