基于人工免疫的网络安全态势评估模型

本文将人工免疫原理应用到网络安全态势评估的研究中.通过对入侵检测模型中有关自体、非自体、抗原的形式化描述,建立了抗体细胞的演化机制、动态免疫记忆、免疫监视等免疫机制.依据人体免疫系统抗体浓度的变化与病原体入侵强度的对应关系,建立了网络安全态势实时定量计算模型.     

模仿呼吸时肺通气试验装置的研制

机械通气是支持和拯救患者生命的重要方式之一。在实行通气临床应用之前,对设备的参数和性能的测定是保证试验更具安全性和可行性的前提。为了建立能够模仿呼吸时肺通气情况的装置来替代活体肺通气作为通气试验对象,以动物肺为肺通气模型核心,采用数据处理和驱动控制,研制了一款通过驱动针筒活塞模仿呼吸肌做功产生吸、呼气流对肺通气模型进行通气的试验装置。通过比较新鲜羊肺为肺通气模型模仿呼吸试验输出的气流波形与实测人体呼吸气流波形,证明了所建试验装置的通气气流波形特征与实测人体呼吸气流波形特征相一致。     

寄生虫病原体自动识别研究

寄生虫病是影响人类健康的重要疾病。诊断该病的最常用和最可靠的方法是检测疑似患者是否携带有寄生虫病原体,然而寄生虫病原体的形态多样性和复杂性及检测人员的主观因素等严重影响了寄生虫病的正确检测。为了改变人工检测方法的低效率,便于医务工作者开展大规模的寄生虫病普查与防治工作,该文利用医学显微图像自动识别技术对人体寄生虫病原体彩色虫卵显微图像的自动识别进行了研究,并且通过对采集到的10种人体寄生虫卵图像进行识别,取得了较好的识别效果。     

寄生虫病原体自动识别研究

寄生虫病是影响人类健康的重要疾病。诊断该病的最常用和最可靠的方法是检测疑似患者是否携带有寄生虫病原体,然而寄生虫病原体的形态多样性和复杂性及检测人员的主观因素等严重影响了寄生虫病的正确检测。为了改变人工检测方法的低效率．便于医务工作者开展大规模的寄生虫病普查与防治工作,该文利用医学显微图像自动识别技术对人体寄生虫病原体彩色虫卵显微图像的自动识别进行了研究,并且通过对采集到的10种人体寄生虫卵图像进行识别,取得了较好的识别效果。     

基于EM算法和曲面拟合模型的肺裂检测算法

肺裂是识别人体肺部解剖结构的重要标记.为解决由于肺裂存在形变、不完整和断裂等现象造成的难以在CT图像中自动检测的问题,提出一种基于期望最大化算法和曲面拟合模型的肺裂检测算法.首先结合气管和动脉血管先验知识估计肺裂区域,减少肺部其他结构的干扰;然后构建肺裂方向场更好的区分肺裂和噪声,并利用期望最大化算法识别肺裂;最后采用曲面拟合模型弥补丢失的肺裂,达到更好地检测肺裂的目的.提出的算法在20个患有慢性阻塞性肺部疾病患者的肺部CT图像上进行实验,结果表明,与人工参考对比,该算法在左肺裂和右肺裂的F_1-score中值分别为0.979和0.971,能够高效率地检测肺裂.     

电荷对粉尘和病原体团聚影响的数学模型研究

由于封闭畜禽舍内聚集了大量的粉尘和病原体,将影响到动物和饲养人员的呼吸道健康,因此迫切需要研究去除粉尘和病原体的方法。以粉尘和病原体的结合体为研究对象,利用蒙特卡罗法及相关概率模型,研究电荷对粉尘和病原体团聚影响。通过研究粉尘和电荷间的附着模型,确定其附着受粒子半径、电荷情况、风速等因素影响;之后,进一步构建粉尘、病原体间以及病原体、电荷间的附着模型。结合数值化运算设定边界条件,得出三者间的结合率可达到15%。粉尘、病原体与电荷相互之间模型的建立能有效的了解到畜禽舍除尘的方法,提高除尘效率,并为高压静电除尘的推广和应用提供了理论基础。     

面对威胁，自动防御

IT基础设施和网络应该像人体一样，人体虽然携带着病素毒和病原体，但人体机能仍可正常运行。自我免疫概念是思科安全构想与战略的构建基础。     

扩展2维环境中的移动机器人多人体目标跟踪

移动机器人如何自主实现人体目标的检测与跟踪是服务机器人研究领域中的关键问题之一.在深入分析单目视觉和激光测距特性的基础上.文章首先针对室内场景进行扩展2维环境建模研究,并提出在该环境下的人体目标分段模型构建与自主辨识方法.为了有效实现对多个人体目标的同时跟踪,本文提出了一种基于非恒速运动模型和卡尔曼滤波对多人体目标进行有效匹配与跟踪的方法.实验表明本文所提方法能有效的克服目标旋转、部分遮挡和重叠以及光线明暗变化给人体目标跟踪带来的影响,具有较好的鲁棒性和实用性.     

人体运动仿真综述

人体运动仿真是生物力学、机器人学和计算机科学交叉产生的研究领域,其目的是建立计算模型模拟人体在给定约束下自然真实的物理运动.与一般的人体动画相比,人体运动仿真要求合成的人体运动满足牛顿定律.因此,人体运动仿真除了可用于特效制作,还可用于物理任务的操作培训和学习等方面.首先介绍人体运动仿真的起源与典型应用,然后阐述人体运动仿真研究的关键技术.在关键技术介绍中,引入人体运动仿真研究中使用的人体简化几何模型和物理模型,然后将人体运动仿真方法分成两类:基于前向动力学方法与基于逆向动力学方法,并详细介绍它们的发展历程和最新研究进展.最后从交互动作的仿真、仿真结果验证等几方面讨论人体运动仿真未来可能的发展方向.     

复杂系统仿真与人体仿真探讨

系统仿真是为复杂系统的组成建立模型,对系统功能的实现进行可视化的研究,是基于系统认识的模型化表示。通过对仿真演示结果评估与分析得到的结论,再与系统模型表达的系统知识进行重合性、再现性比较研究。在生理系统中人体系统仿真研究分为医学人体仿真研究和应用人体仿真研究。由人体切片数据重建的人体构造三维图像,是在计算机虚拟空间中对人体解剖数据的可视化;下一阶段的医学人体仿真研究工作是建立各种器官系统的附加相应的物理性、生理性等参数的仿真模型模块,在仿真集成环境中进行仿真,并对仿真研究结果加以评估和分析。     

基于仿生微流控技术的肠道器官芯片构建

人体肠道具有复杂的生理环境,存在肠上皮细胞形成的绒毛结构以及流体剪切力、肠道蠕动 等力学条件,且肠道中与人体共生着大量微生物,这些与人类的健康息息相关。对于肠道的研究,传 统的体外细胞培养手段不能完全模拟肠道复杂的生理环境,具有局限性。该研究设计了一种基于微流 控技术的肠道器官芯片,通过将肠细胞培养于绒毛状基膜,结合流体剪切力来模拟人体肠道的结构与 功能。结果显示,该肠芯片重现了肠道的绒毛结构与屏障功能,提高了黏液的表达量,并实现了肠道 菌在肠芯片上的实时观察,可以成为在体外研究肠道微生物与宿主相互作用的有力工具。     

基于机器学习的肺音分类技术的研究进展

肺音(Lung Sound) 信号是人体呼吸系统与外界在换气过程中产生的一种生理声信号,其因含有大量的生理和病理信息而具有很高的研究价值。近年来,频发的雾霾天气等环境问题所带来的呼吸道疾病发病率的提高,也使得对肺部疾病诊断的快速性与准确性的需求大幅提升。肺部听诊以其迅捷便利和无创等优良特性重新引发人们的广泛关注,而自动肺音诊断技术的发展无疑会对肺部疾病诊断带来重要的帮助。电子听诊器以及其他信号采集技术等硬件方面的发展进一步促进了现代肺音信号的分析和识别技术的研究与进步。主要介绍了肺音的概念、基于计算机的肺音信号处理和模式识别技术,并对近年来基于机器学习的肺音分类技术的发展状况进行了总结与列举;最后,对肺音分类技术的研究和应用发展趋势进行了展望。     

肺癌呼吸气体检测系统的气室设计

基于卟啉传感器阵列(PSA)的肺癌呼吸气体检测系统是通过采集处理PSA芯片表面图谱信息来对人体呼吸气体进行定性检测,达到早期肺癌诊断的目的。检测过程中,PSA芯片能否与气体分子快速、均匀结合将直接影响气体检测系统的准确度。以系统中PSA芯片与肺癌呼吸气体反应的气室为研究对象,借助CFD专用软件FLUENT对气室进行流场仿真结构优化,在气室理论模型的基础上,进行了气室实物制作,结果表明:通过采用抛物线形气室内腔、安装导流板、减小分流孔通径、适当加大分流孔倾斜角度等优化方案后,气室内呼吸气体呈层流流动状态,流场更加均匀、密集,能达到检测系统中PSA芯片与呼吸气体分子快速、均匀结合的要求。     

Challenges and opportunities for cell line secretomes in cancer proteomics

Cancer cell lines are the most widely used experimental models in cancer research. Their advantages of easy growth and manipulation are unfortunately paralleled by their limitations derived from long-term growth in isolation from the rest of the tumor, and hence, lack of tumor microenvironment. We are however currently witnessing novel and transformative advances that are making cell lines more reflective of the human biology and therefore, better experimental models for cancer research. Beyond the experimental model used, the choice of cellular proteome is key in proteomics-based biomarker discovery. Over the last decade, cell line secretomes have been proposed as an alternative for tumor biomarker discovery due to the difficulties posed by plasma in terms of complexity and low abundance of tumor-specific biomarkers. Cell line secretomes are enriched with proteins already linked to tumorigenesis, which also have a good chance of being present in biological fluids. In this review, we will provide an overview of the main technical and biological issues related to cell line secretome analysis, and briefly discuss both the challenges and opportunities in its use for tumor biomarker discovery.     

利用肌电信号求解关节力矩的研究及应用综述

表面肌电信号(surface electromyography, sEMG)是人体的易于检测的神经信号,其富含大量人体运动信息。利用肌电信号作为输入信号,结合相关生物学模型分析肌电信号同肌肉力和对应关节力矩之间的关系,对于深入理解分析人体动力学具有重要意义。本文详细归纳总结了利用肌电信号求解人体关节力矩方法的研究成果,同时介绍神经肌肉骨骼模型的计算及优化过程,给出部分模型生理参数为之后的研究提供参照,并给出现阶段该方法在人体关节力矩求解中的应用。通过分析该求解过程中所面临的一些问题,总结出该方法的发展展望,为之后的研究提供参考。     

Mathematical analysis of oxygen transfer through polydimethylsiloxane membrane between double layers of cell culture channel and gas chamber in microfluidic oxygenator

For successful cell culture in microfluidic devices, precise control of the microenvironment, including gas transfer between the cells and the surrounding medium, is exceptionally important. The work is motivated by a polydimethylsiloxane (PDMS) microfluidic oxygenator chip for mammalian cell culture suggesting that the speed of the oxygen transfer may vary depending on the thickness of a PDMS membrane or the height of a fluid channel. In this paper, a model is presented to describe the oxygen transfer dynamics in the PDMS microfluidic oxygenator chip for mammalian cell culture. Theoretical studies were carried out to evaluate the oxygen profile within the multilayer device, consisting of a gas reservoir, a PDMS membrane, a fluid channel containing growth media, and a cell culture layer. The corresponding semi-analytical solution was derived to evaluate dissolved oxygen concentration within the heterogeneous materials, and was found to be in good agreement with the numerical solution. In addition, a separate analytical solution was obtained to investigate the oxygen pressure drop (OPD) along the cell layer due to oxygen uptake of cells, with experimental validation of the OPD model carried out using human umbilical vein endothelial cells cultured in a PDMS microfluidic oxygenator. Within the theoretical framework, the effects of several microfluidic oxygenator design parameters were studied, including cell type and critical device dimensions.     

骨小梁仿生微结构的解析与构建

骨小梁的生物机制因其复杂的三维微结构和微环境及其与成骨细胞之间的相互作用,已被世界各国学者在骨修复重建和骨再生领域中研究和应用。然而,以往的研究缺乏对体内骨小梁真实环境的模拟,尤其是缺乏对骨小梁微结构的精确量化及其与微环境相互作用条件下骨骼再生的研究与应用。该研究开发一种新的骨小梁仿生微结构的解析与构建方法来分析和构建骨小梁的三维环境。该方法基于 Micro CT 兔股骨近端松质骨扫描,采用 Mimics 和 Rhino 软件构建三维松质骨模型并分析其几何特征,应用斯特林公式和 3D 打印骨小梁结构,对仿生骨小梁结构进行了解析和构建。截面积对比和结构相似性分析表明,仿生松质骨与天然松质骨结构达到很高的相似度(达 95% 以上);体外细胞实验表明,仿生骨小梁微环境能够提供较好的细胞生长微环境。该研究可为组织工程临床研究与应用提供有力的工具。     

Computational fluid dynamics of nanoparticle disposition in the airways: mucus interactions and mucociliary clearance

Interactions of nanoparticles with respiratory fluids such as airway mucus are currently under investigation and are involved in a variety of applications. The clearance processes of those nanoparticles are still not fully understood. This study presents an approach to describe deposition, sedimentation and clearance of nanoparticles within mucus with numerical and analytical models: Particle deposition as well as motility, sedimentation and clearance were simulated with Computational Fluid Dynamics (CFD) and described analytically. Furthermore mucus plasticity as pathway for complex particle translocation was simulated using grid-free CFD methods. We could demonstrate that fluid dynamics strongly influence the fate of deposited nanoparticles in mucus: Sedimentation, impaction and diffusion were shown to be unlikely to contribute to particle translocation. However, intrinsic plasticity of mucus slabs and collision of such slabs may enhance particle translocation towards the pulmonary epithelium.     

A mathematical model for foreign body reactions in 2D

The foreign body reactions are commonly referred to the network of immune and inflammatory reactions of human or animals to foreign objects placed in tissues. They are basic biological processes, and are also highly relevant to bioengineering applications in implants, as fibrotic tissue formations surrounding medical implants have been found to substantially reduce the effectiveness of devices. Despite of intensive research on determining the mechanisms governing such complex responses, few mechanistic mathematical models have been developed to study such foreign body reactions. This study focuses on a kinetics-based predictive tool in order to analyze outcomes of multiple interactive complex reactions of various cells/proteins and biochemical processes and to understand transient behavior during the entire period (up to several months). A computational model in two spatial dimensions is constructed to investigate the time dynamics as well as spatial variation of foreign body reaction kinetics. The simulation results have been consistent with experimental data and the model can facilitate quantitative insights for study of foreign body reaction process in general.     

Numerical study on the effect of nerve control on upper airway collapse in obstructive sleep apnea

Obstructive sleep apnea syndrome (OSAS) is a respiratory disease characterized by the upper airway collapses and reopens repeatedly during sleep. Though the nerve control plays a key role in the upper airway collapse, it has been considered in previous studies only with lumped parameter models. Based on a finite element model including airway and surrounding structures, the effect of nerve control on the upper airway collapse was studied with fluid-structure interaction method. Spring elements were used to simulate the function of the muscle group. The simulation results show that the nerve control reduces the deformation of airway successfully and avoids the risk of OSAS.