功能性磁振造影網路資料庫系統之開發：文獻探勘技術之初步研究

Development of a Web-Based Functional Magnetic Resonance Imaging Database System: A Preliminary Study of Text Mining Technique

易睿祥(Ray-Shawn Yih)

指導教授 ： 陳志宏

國立臺灣大學/電機資訊學院/電機工程學研究所/碩士(2006年)

https://doi.org/10.6342/NTU.2006.02891

摘要

神經認知科學是一門蓬勃發展中的學問，加上MRI技術近年來的愈益成熟，關於人腦fMRI的相關研究論文已累積有十萬篇以上，面對如此龐大的文獻資料，需要有系統的彙整管理並提供更進一步的資訊。
所以本論文建立了一個網路資料庫系統，透過顯示大腦解剖區域及反應活化區的影像，與提供fMRI實驗與相關文獻查詢資訊，讓認知心理學家在使用本系統檢索影像的過程中，可以即時獲得與該實驗及反應區相關的文獻資訊，解決了認知心理學家對於龐大資料管理的麻煩，以及需要配合其他外掛程式才能查詢解剖區域名稱，無法即時與其他認知心理學家分析的結果比較，和缺少立即相關文獻資訊回顧的輔助等問題。
本資料庫系統，可以透過對大腦各區域活化程度的統計分析，深入瞭解反應區的分佈情形，並且提供兩兩實驗之間的交叉比對功能，來觀察大腦反應活化的交集與聯集狀況，搭配顯示相關文獻摘要的服務，將可使神經認知科學之發展更為深入且有效率。

關鍵字

功能性磁振造影 ； 網路資料庫系統

並列摘要

Cognitive neuroscience has been studied for several years. For the last decade, the technique of MRI has already become the famous tool for the research of cognition neuroscience. Over than 0.1 million journals have been published about human fMRI. Therefore, the development of database is necessary for managing a large number of correlated references and experiment datasets.
The database system of this thesis has the capability of illustrating brain anatomical labels, retrieving fMRI experiments, and providing corresponding text information. That helps psychologists manage large quantities of data, compare analyzed fMRI results with other researchers, and no longer need to install any plug-in programs to query anatomical labels information.
The system is not only able to provide the statistical analysis of each anatomical label but also make users to figure out the distribution of activation regions clearly. Furthermore, by comparing with two experiments, users can also observe the union and intersection situation conveniently. Finally, the function of displaying corresponding text mining information would be provided by this system, and it would promote cognitive neuroscientists to study fMRI in depth and more efficiently.

並列關鍵字

Functional Magnetic Resonance Imaging ； Web-Based Database System

參考文獻

[2] Barinaga M., Neuroimaging, Still debated, brain image archives are catching on. Science. 2003 Apr 4;300(5616):43-5.

[3] The Statistical Parametric Mapping.

[4] Editorial. A debate over fMRI data sharing. Nature Neurosci. 3, 845-846 (2000).

[5] OHBM, Neuroimaging Databases, SCIENCE, JUNE 2001, VOL292

[8] The Whole Brain Atlas