脑微观重建系统研究

发布时间：2018-06-26 02:09

本文选题：脑计划 + 神经回路重建　；参考：《系统工程理论与实践》2017年11期

【摘要】：本文旨在研究如何利用硬件,软件,算法的系统工程的方案解决脑微观结构重建的高通量自动化数据分析的难题.通过实现样片制备,自动切片,显微成像,三维重建以及软件平台等分阶段工作,建立符合神经结构生物学特征的模式识别和机器学习分类方法,解决海量畸变图像拼接配准,密集神经结构识别,歧义神经联结,多源数据融合等限制分析效率的关键问题,为搭建大体量神经结构重建工程平台提供理论基础和解决方案,满足脑科学研究对高通量神经回路网络重建的需求.
[Abstract]:The purpose of this paper is to study how to solve the problem of high-throughput automated data analysis of brain microstructure reconstruction by using hardware, software and algorithm of system engineering. In order to solve the problem of large amount of distorted image stitching registration, pattern recognition and machine learning classification method, which accord with the biological characteristics of neural structure, are established through the phased work of sample preparation, automatic slicing, microscopic imaging, 3D reconstruction and software platform. The key problems which limit the efficiency of analysis, such as the identification of dense nerve structure, the connection of ambiguous nerve and the fusion of multi-source data, provide the theoretical basis and solution for the construction of a large volume neural structure reconstruction engineering platform. To meet the needs of high throughput neural network reconstruction in brain science.
【作者单位】：北京工业大学经济与管理学院;中国科学院自动化研究所;中国科学院脑科学与智能技术卓越创新中心;中国科学院大学未来技术学院;
【基金】：国家自然科学基金(61673381,61201050,61306070) 北京市科委项目(2161100000216146) 中国科学院先导项目(X DB02060001);中国科学院自动化研究所突触级三维重建项目(Y3J2031DZ1);中国科学院科研仪器设备研制项目(YZ201671)~~
【分类号】：R318;TP391.41

【相似文献】