权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Development of an automated somatic embryo production system equipped with image analysis facilities

开发配备图像分析设施的自动化体细胞胚胎生产系统

基本信息

批准号：
03556036
负责人：
KURATA Kenji
金额：
$ 10.62万
依托单位：
The University of Tokyo
依托单位国家：
日本
项目类别：
Grant-in-Aid for Developmental Scientific Research (B)
财政年份：
1991
资助国家：
日本
起止时间：
1991 至 1993
项目状态：
已结题

项目摘要

The objective of the study was to develop an automated somatic embryo production system, in which good quality torpedo stage embryos were selected for further treatment of transplant or artificial seed production. The selection was executed based on image analysis of each somatic embryo. The study consisted of three parts.1. Development of an automated embryogenic callus selection system. In a conventional manual method of somatic embryo production, embryogenic callus was selected by sieving with three different size mesh. This procedure was automated and tested using carrot callus. The results showed increased ratio of malformed embryos by mechanical sieving, but that the automation was possible.2. Development of image analysis method of somatic embryos. Image analysis algorithms for selecting quality torpedo stage embryos were developed and tested. Algorithms based on thinning, generalized Hough transform, and Fourier descriptor were compared using carrot somatic embryo images. The results showed that the thinning algorithm was most promising which gave results 90% in agreement with human judgments.3. Development of SOMES (SOMatic Embryo Sorter). A system for selecting embryos was constructed and tested. The thinning algorithm was implemented. Path of liquid medium containing embryos was changed according to the results of image analysis of each embryo. The results showed the effectiveness of SOMES, but at the same time difficulties in image analysis of moving embryos in the imaging cell. The reasons of misjudgments were analyzed and points to be improved were suggested : use of a thinner tube for the imaging cell, and acquiring images from two angles. Efforts are being made to improve SOMES in this direction.

本研究的目的是开发一个自动化的体细胞胚胎生产系统，在该系统中，选择质量好的鱼雷期胚胎进行移植或人工种子生产的进一步处理。基于每个体细胞胚的图像分析进行选择。本研究分为三个部分.胚性愈伤组织自动筛选系统之建立。在体细胞胚生产的常规手工方法中，通过用三种不同尺寸的筛网筛选胚性愈伤组织。该过程是自动化的，并使用胡萝卜愈伤组织进行测试。结果表明，机械筛选会增加畸形胚的比例，但自动化筛选是可行的.体细胞胚图像分析方法的建立。开发并测试了用于选择优质鱼雷阶段胚胎的图像分析算法。以胡萝卜体细胞胚图像为研究对象，对基于细化、广义Hough变换和傅立叶描述子的算法进行了比较。结果表明，细化算法最有前途，其结果与人类判断的一致率达到90%. SOMES（体细胞胚胎分选仪）建立了一个胚胎筛选系统并进行了试验。实现了细化算法。根据每个胚胎的图像分析结果改变含有胚胎的液体培养基的路径。结果显示了SOMES的有效性，但同时也存在成像细胞中运动胚胎图像分析的困难。分析了造成误判的原因，提出了改进的方向：采用更细的成像管，从两个角度获取图像。目前正在努力朝这个方向改进妇女和儿童教育。