基于Matlab GUI的脑部核磁共振数据统计分析系统

发布时间：2018-05-03 16:51

本文选题：Matlab + GUI　；参考：《电子科技大学》2013年硕士论文

【摘要】：近年来，随着磁共振成像技术的迅速发展以及产品成本的下降，核磁共振成像仪开始在各大医院及科研院校被逐步推广。除临床疾病诊断所需的检查外，使用磁共振成像技术探测人脑功能以及结构也是当今神经科学研究的热点。对大脑功能及结构磁共振数据的研究不仅要求具备相关的物理学基础，还要掌握必要的医学知识，同时也需精通信号处理分析技术，这种多学科的交叉融合对研究者的科研素养提出了极高要求。基于此，本文采用基于Matlab GUI的界面设计，发展了一个脑部磁共振数据统计分析系统。此系统涵盖了从原始数据整理到图像统计分析再到结果图像输出的一套完整流程。结合实际科研工作的需要，此系统采用的数据处理分析方法均为当今科学界公认的可靠而又先进的方法。同时系统采用基于Matlab GUI的界面设计，，其操作直观简便，让使用者极易掌握。此系统大致分为以下几个模块： 1. DICOM数据整理。根据头信息(.hdr）批量将DICOM文件改名,同时按扫描序列、日期、被试类型等进行归类。 2. fMRI数据处理。对功能磁共振数据进行感兴趣区（ROI）信号提取、全脑相关网络、网络间相关、功能连接密度等分析。 3.结构MRI数据处理。对DTI数据进行纤维束微结构信息获取、全脑纤维束连接矩阵建立、网络属性计算、纤维束方向密度函数（ODF）图分析等处理。 4.小工具。包括低频成份频谱分析、种子点优秀程度判断、全脑模板分割、去除模板散点等实用功能。通过编排合理的界面将上述功能整合，形成了一套界面简洁、操作方便、可靠度高的适用于科研的人脑MRI数据统计分析系统，具有较高的推广价值。
[Abstract]:In recent years, with the rapid development of magnetic resonance imaging technology and the decline of product cost, nuclear magnetic resonance imager has been gradually popularized in various hospitals and scientific research institutions. In addition to the necessary examinations for the diagnosis of clinical diseases, magnetic resonance imaging (MRI) is also a hot topic in neuroscience. The study of functional and structural magnetic resonance data of the brain requires not only the relevant physical basis, but also the necessary medical knowledge, as well as a good command of signal processing and analysis techniques. This kind of multi-disciplinary cross-fusion has put forward the extremely high request to the researcher's scientific research accomplishment. Based on this, a statistical analysis system of brain magnetic resonance data is developed by using interface design based on Matlab GUI. The system covers a complete process from raw data collation, image statistical analysis to result image output. According to the need of scientific research, the data processing and analysis methods used in this system are recognized by the scientific community as reliable and advanced methods. At the same time, the system adopts interface design based on Matlab GUI, its operation is intuitionistic and simple, which makes the user easy to master. The system is broadly divided into the following modules: 1. DICOM data collation. The DICOM file was renamed in batches according to the header information, and classified according to the scanning sequence, date and type of the subject. 2. FMRI data processing The data of functional magnetic resonance (fMRI) were extracted from ROI, the whole brain correlation network, the correlation between networks and the density of functional connection were analyzed. 3. Structure MRI data processing. The DTI data were obtained from the microstructures of the fiber bundle, the whole brain fiber bundle connection matrix was established, the network attribute was calculated, and the fiber bundle direction density function (DDF) graph was analyzed. 4. Gadget. It includes low frequency component spectrum analysis, seed point excellent degree judgment, whole brain template segmentation, removal of template scattered points and other practical functions. By arranging a reasonable interface to integrate the above functions, a set of human brain MRI data statistical analysis system with simple interface, convenient operation and high reliability is formed, which has high popularizing value.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP391.41;R310

【参考文献】