應用多頻譜影像再現技術於辨識膀胱癌細胞分期之研究

Study of multispectral image reproduction for diagnosis of bladder cancer with different stages

林聰智(Lin, Tsung Chih)

指導教授 ： 王祥辰

國立中正大學/工學院/光機電整合工程研究所/碩士(2012年)

摘要

本論文提出一種應用多頻譜影像再現技術於辨識膀胱癌細胞分期之方法，研究內容分為兩個部份：第一部份為利用多頻譜影像系統分析模擬的膀胱細胞頻譜；第二部份則應用主成分分析圖將不同分期的膀胱細胞頻譜做分類。本系統為利用多頻譜影像技術以及色彩再現並搭配主程分分析圖實現本研究。實驗結果表明，在主程分分析圖分類裡，使用第一主成分以及第二主成分將比第一主成分以及的三主成分更符合需求。根據這項特點，經由主成分分佈點的運算，我們可以將正常膀胱細胞(E7)、第二期膀胱癌(TSGH-8301)、第三期膀胱癌(J82)以及第四期膀胱癌(TCC-SUP)以橢圓的方程式分成三類，分別為正常、(第2、3期)以及第四期。未來我們希望能夠利用此技術，有效地提升醫生診斷效率，幫助病患進行早期治療。

關鍵字

多頻譜 ； 色彩再現 ； 膀胱癌

並列摘要

The multispectral image reproduction for diagnosis of bladder cancer with different stages is proposed. The study can be divided into two parts: the investigation of multispectral system, which simulate the spectra of bladder cell, and the research of score plot, which categorize bladder cell with different stages. A system combined with multi-spectral imaging (MSI), color reproduction technique and score plot is applied to achieve the study. The results show that using the first principle component and the second principle component are more satisfying than the first principle component and the third principle component in analyzing of socre plot. According to this, we can categorize bladder cell line (E7), bladder cell line (TSHG-8301), bladder cell line (J82) and bladder cell line (TCC-SUP) into three groups using ellipse function, which can divide these bladder cell line into normal, stage2,3 and stage 4. In the future, we hope we can make medical diagnosis more efficiency and help patients for early treatment applying our technique.

並列關鍵字

multispectral ； color reproduction ； bladder cancer

參考文獻

[31] H.G. van der Poel , M.E. Boon, E.A. van der Meulen , and A. Wijsman- Grootendorst, “The reproducibility of eytomorphometrieal grading of

[1] National Cancer Institute

[2] World Health Organization

[3] Bureau of Health Promotion, Department of Health, R.O.C.

[4] J.K. Kim, S. Y. Park, H. J. Ahn, C . S. Kim, K. S. Cho, “Bladder Cancer: Analysis of Multi–Detector Row Helical CT Enhancement Pattern and Accuracy in Tumor Detection and Perivesical Staging,” Radiology, 231, 725-731 (2004).