• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Conference on Optical Diagnostics of Living Cells Ⅴ, Jan 23-25, 2002, San Jose, USA >Tracking of tissue section surfaces for automated 3D confocal cytometry
【24h】

Tracking of tissue section surfaces for automated 3D confocal cytometry

机译：跟踪组织切片表面以进行自动3D共聚焦细胞计数

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Three-dimensional cytometry, whereby large volumes of tissue would be measured automatically, requires a computerized method for detecting the upper and lower tissue boundaries. In conventional confocal microscopy, the user interactively sets limits for axial scanning for each field-of-view. Biological specimens vary in section thickness, thereby driving the requirement for setting vertical scan limits. Limits could be set arbitrarily large to ensure the entire tissue is scanned, but automatic surface identification would eliminate storing undue numbers of empty optical sections and forms the basis for incorporating lateral microscope stage motion to collect unlimited numbers of stacks. This walkaway automation of 3D confocal scanning for biological imaging is the first step towards practical, computerized statistical sampling from arbitrarily large tissue volumes. Preliminary results for automatic tissue surface tracking were obtained for phase-contrast microscopy by measuring focus sharpness (previously used for high-speed autofocus by our group). Measurements were taken from 5x5 fields-of-view from hamster liver sections, varying from five to twenty microns in thickness, then smoothed to lessen variations of in-focus information at each axial position. Because image sharpness (as the power of high spatial frequency components) drops across the axial boundaries of a tissue section, mathematical quantities including the full-width at half-maximum, extrema in the first derivative, and second derivative were used to locate the proximal and distal surfaces of a tissue. Results from these tests were evaluated against manual (i.e., visual) determination of section boundaries.
机译：三维流式细胞术可自动测量大量组织，因此需要一种用于检测组织上下边界的计算机化方法。在传统的共聚焦显微镜中，用户可以交互式设置每个视野的轴向扫描限制。生物样本的切片厚度各不相同，从而推动了对垂直扫描范围设置的要求。可以将限制设置为任意大的值，以确保扫描整个组织，但是自动表面识别将消除存储过多数量的空光学部分，并为结合横向显微镜载物台运动以收集无限数量的堆叠奠定了基础。这种针对生物成像的3D共聚焦扫描的自动化实现是朝着从任意大组织量进行实用计算机统计采样的第一步。通过测量聚焦锐度（我们小组以前用于高速自动聚焦），通过相衬显微镜获得了组织表面自动跟踪的初步结果。从仓鼠肝脏切片的5x5视野中进行测量，厚度从5到20微米不等，然后进行平滑处理以减少每个轴向位置的对焦信息变化。由于图像清晰度（作为高空间频率分量的幂）在组织切片的轴向边界上下降，因此使用数学量（包括半峰全宽，一阶导数和二阶导数的极值）来定位近端和组织的远端表面。将这些测试的结果与手动（即视觉）确定截面边界进行了评估。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号