透過您的圖書館登入
IP:18.119.131.72
  • 學位論文

擴散光學斷層掃描術二維影像重建之模擬分析

Simulation and analysis of two-dimensional image reconstruction in diffuse optical tomography

指導教授 : 蔡睿哲
共同指導教授 : 孫家偉

摘要


在我們人體組織裡,對於波長約在600-900 nm近紅外光的吸收係數較小,光在組織中也能穿透較深。因此,善於利用此特性可以讓非侵入式的光學(醫療)儀器達到可穿透的最深深度,以一窺體內部份特徵。 擴散光學斷層掃瞄術便是利用此波段的近紅外光為醫學影像檢測光源,利用這些光子的強度與分布等資訊可以重建出在生物組織內吸收及散射隨空間與時間分布的變化。在目前的臨床應用上,是以胸部腫瘤、大腦功能性造影、以及運動醫學為主要研究探討的對象,在未來也是以這些為主軸進而在發展出應用在更廣泛的層面上。此技術的優點是非侵入式、即時檢測、非輻射源、功能性生理量測。 一般來說,擴散光學斷層掃瞄術的開發可大分為兩項,一是在系統上檢測光源的使用以及光偵測器的靈敏度等,二是在重建二維甚至於三維影像的演算法,影像的品質取決於演算法的使用。在此篇論文裡,我們模擬在紊亂介質中各種幾何形狀的非均勻吸收體來重建影像,並且使用數學上的同步疊代法,為了在系統運算時間與影像品質上取得平衡點,疊代過程需要一個最佳停止點。 從模擬的結果裡,我們證實了這些演算法適用於擴散光學斷層掃瞄術並且能有效地重建影像。

並列摘要


The near-infrared light is weakly absorbed between 600-900 nm in the human tissue. Therefore, the optical imaging system with the deep penetration characteristic can be applied to biomedical instrument. Diffuse optical tomography (DOT) is an emerging technique for biomedical imaging based on the near-infrared photons. It is a technique that images spatial variations in absorption and scattering properties of the biological tissues. For the clinical studies, the breast cancer detection, the functional brain imaging, and the diagnosis of exercise medicine will be concerned and focused in further research. It has the advantages of noninvasive, real-time, non-radiation and functional imaging. Generally, the categories of DOT development are divided into two parts: one is the devices choice of light source and photo-detector and the other is the image reconstruction algorithm. The image quality of DOT technique strongly depends on the reconstruction algorithm. In this thesis, inhomogeneities with various shapes of absorption distributions in turbid media are simulated. The DOT images are reconstructed based on the simultaneous iterative reconstruction technique (SIRT) method. To solve the trade-off problem between time consumption of reconstruction process and accuracy of reconstructed image, the iteration process needs an optimization criterion in algorithm. The simulation results reveal that the SIRT algorithm can offer efficient image reconstructing in DOT system.

參考文獻


1-6 D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M.
tomography,” Science 254, 1178-1181 (1991).
1-8 D. A. Boas, A. M. Dale, and M. A. Franceschini, “Diffuse optical imaging of
brain activation: Approaches to optimizing image sensitivity, resolution and
accuracy,” NeuroImage 23, S275–S288 (2004).

延伸閱讀