CASIA OpenIR  > 毕业生  > 硕士学位论文
高空间分辨率的功能近红外断层成像技术研究
Alternative TitleResearch in High-resolution Functional Near-infrared Tomography
吉利军
Subtype工学硕士
Thesis Advisor蒋田仔
2007-01-26
Degree Grantor中国科学院研究生院
Place of Conferral中国科学院自动化研究所
Degree Discipline模式识别与智能系统
Keyword功能近红外断层成像 功能近红外光谱 多中心光纤排布模式 图像质量 空间分辨率 光源-接收器对 Functional Near-infrared Tomography Functional Near-infrared Spectroscopy Multi-centered Mode Image Quality Spatial Resolution Source-detector Pair
Abstract功能近红外断层成像技术是一种无创的大脑皮层功能活动检测手段。相比于其它的大脑皮层功能活动检测方法,它具有较高的时间分辨率,低廉的设备成本和数据采集成本,可以实时的看到结果,是一种具有巨大潜力的技术。然而,功能近红外断层成像技术较低的空间分辨率导致图像质量较差,限制了它的发展。 本文通过对功能近红外断层成像技术原理的分析,总结出了要提高空间分辨率和图像质量,必须提高采集数据中有效信息的含量;要提高数据质量,需要更多的光源-接收器对,并使每个象素都有多个光源-接收器对检测;要提高光源-接收器对的数量,必须改善光纤排布模式的理论。根据这一理论,本文发明了“多中心光纤排布模式”。多中心光纤排布模式有如下的优点: 1、多中心光纤排布模式可以得到168个光源-接收器对,远高于现有的光纤排布模式; 2、在多中心光纤排布模式的中心附近区域,每个象素平均有35.5个光源-接收器对可以检测,在周边区域,每个象素平均有11.7个光源-接收器对可以检测,远高于现有光纤排布模式; 3、多中心光纤排布模式中的光源在实验中可以同时打开,相比于之前的光源分组打开模式,多中心光纤排布模式可以为大大提高实验的便利性; 4、在半无限模型中的仿真实验证明多中心光纤排布模式可以得到比之前的排布模式更高质量的图像。 通过本文的描述我们可以得出结论:功能近红外断层成像技术可以具有很高的图像质量,可以具有很高的空间分辨率。 本文的后续工作主要包括在人体实验中进一步验证多中心光纤排布模式的性能、设计覆盖整个大脑皮层的高分辨率光纤排布模式等。
Other AbstractFunctional near-infrared tomography (fNIT) is a new tool for detecting of human brain activations. Comparing to other functional imaging methods, fNIT has better temporal resolution, low instrumental and experimental costs, and Real-time display of images. Unfortunately, the poor image quality which caused by the poor spatial resolution has confined its clinical applications. This paper has achieved that, the key to improve spatial resolution and image quality is to obtain more collected data; the key to obtain more collected data is to get more source-detector pairs and more overlapping of source-detector pairs; the key to get more source-detector pairs is to employ better arranging modes of optical fibers. This paper has developed multi-centered arranging mode of optical fibers, which has advantages in following aspects. 1. The multi-centered mode can achieve 168 source-detector pairs, and exceeds the previous modes much. 2. Each pixel around the central region of the multi-centered mode can be detected by averaging 35.5 source-detector pairs. And each pixel around the peripheral region of the multi-centered mode can be detected by averaging 11.7 source-detector pairs. 3. The sources in the multi-centered mode can be simultaneously turned on during the experiments. However, the sources in previous modes must be turned on sets by sets to avoid saturation of detectors, so the multi-centered mode can make fNIT experiments more convenient. 4. Simulations in a semi-infinite model show that higher-quality images can be obtained comparing to the previous modes. This paper concludes that fNIT can achieve high-quality images and have good spatial resolution. The future extends of this paper will lay on proving the multi-centered mode in people experiments and designing further modes for covering whole brain.
shelfnumXWLW1053
Other Identifier200328014604128
Language中文
Document Type学位论文
Identifierhttp://ir.ia.ac.cn/handle/173211/7396
Collection毕业生_硕士学位论文
Recommended Citation
GB/T 7714
吉利军. 高空间分辨率的功能近红外断层成像技术研究[D]. 中国科学院自动化研究所. 中国科学院研究生院,2007.
Files in This Item:
File Name/Size DocType Version Access License
CASIA_20032801460412(1596KB) 暂不开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[吉利军]'s Articles
Baidu academic
Similar articles in Baidu academic
[吉利军]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[吉利军]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.