請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/67141
標題: | 利用超音波影像評估經放射線治療後頭頸部癌症病患肌肉纖維化之量化研究 Quantitative Evaluation of Fibrotic Change in the Irradiated Muscle of Head and Neck Cancer Patients Using Ultrasonography |
作者: | Jhe-Yuan Guo 郭哲源 |
指導教授: | 林文澧(Win-Li Lin) |
關鍵字: | 超音波,頭頸部癌症,放射線引起組織纖維化,Frangi filter,Fuzzy c-means群集演算法,剪力波, ultrasound,head and neck cancer,radiation-induced fibrosis,Frangi filter,fuzzy c-means clustering,shear wave, |
出版年 : | 2017 |
學位: | 碩士 |
摘要: | 背景:
頭頸部癌症病患一般需要接受放射線治療,其副作用可能會在數月或數年後發生,例如:牙關緊閉、吞嚥困難、頸部肌肉僵硬化(Sclerosis)、移動受限和肌肉萎縮等,因而需要定期回醫院進行追蹤,醫生通常會使用觸診和超音波影像查看。然而,受限於醫師的經驗及無客觀的分析數據供其他醫師參考,恐怕會導致檢測失準,因而,若有客觀的影像分析,將會對醫師在復健療程設計上有很大的幫助。除此之外,超音波屬於低傷害風險,對於觀察肌肉結構有良好的空間解析度,也能產生剪力波來探討組織的彈性。 目的: 透過超音波影像建立客觀有效的肌肉纖維化評估系統,以利於未來醫師在復健療程設計與疾病的研究。 材料與方法: 本研究共有27位頭頸部癌症病患,其中有19位男性和8位女性,年齡介於25~65歲之間,並接受66~70Gy劑量放射線治療後,持續1年的治療後追蹤,包含肌束橫切B型超音波影像和縱向剪力波波速量測。本研究將每位病人的胸鎖乳突肌橫切超音波影像,經由Frangi filter和Fuzzy c-means群集演算法分析,得到胸鎖乳突肌的肌肉纖維比,以評估肌肉纖維化程度。 結果: 研究結果顯示在低對比度區域,此方法能有效偵測纖維化,放射線治療後三個月纖維增量有顯著性上升,與一年後胸鎖乳突肌的面積減少有關(r = 0.3571, p-value <0.05),而50歲以上的病人有更高度的相關性(r = 0.6024, p-value <0.05)。本研究發現,在於放射線導致的肌肉纖維化,會在放射線治療後三個月時,其影像灰階值和纖維化增量有顯著性上升,然而其數值卻會在放射線治療後一年下降。 結論: 本研究所研發的纖維偵測演算法,能客觀且定量化分析胸鎖乳突肌放射線治療後引起的纖維化,且有機會在放射線治療後三個月的纖維增量,進而預測一年後的肌肉萎縮情形,未來可提供臨床醫師作為療程設計上的參考依據。 Background: The patients with head and neck cancers generally receive radiation treatment. Radiation-induced toxicity could lead to long-term progressive sequelae, such as trismus, dysphagia, muscle sclerosis, restriction of motion range and atrophy. These patients need long-term follow-ups routinely with palpation and/or ultrasonography, which are all subjective and qualitative diagnosis. Objective and quantitative analysis is helpful for doctors to study clinical manifestations, optimize treatment strategies, and predict the prognosis after radiation treatment. Ultrasound is used to demonstrate muscle structure owing to high spatial resolution. In recent years, shear wave elastography which generated by ultrasound impulse was applied in muscle structure analysis. Material and methods: In this study, 19 males and 8 females of age between 25 to 65 years old were involved, and they received 66 to 70 Gy radiation dose during treatment. We acquired B-mode ultrasonography and measured the velocities of shear wave during one year follow-ups. Frangi filter and Fuzzy c-means clustering algorithm were applied to segment fibrotic tissue from the B-mode images of sternocleidomastoid muscle (SCM). The ratio of fiber to muscle was calculated to quantify fibrosis of muscle. Results: The methods in this study extracted fibrotic tissue from low-contrast region. Fiber increment of SCM increased significantly at 3 months after radiotherapy (RT). Correlation between the fiber increment at 3 months after RT and the area decrement of SCM at one year after RT is positive(r = 0.3571, p-value<0.05), and there is a significant trend (r = 0.6024, p-value <0.05) for the patients older than 50 years. The results showed that radiation-induced muscle fibrosis increased at 3 months after RT, but it declined at one year after RT. Conclusion: The proposed methods quantified the radiation-induced fibrosis of SCM. There was a correlation between the fiber increment of SCM at 3 months after RT and the area decrement of SCM at one year after RT. They may support clinical doctors to analyze the radiation fibrosis syndrome and may be used to evaluate the treatment plan for rehabilitation. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/67141 |
DOI: | 10.6342/NTU201702969 |
全文授權: | 有償授權 |
顯示於系所單位: | 醫學工程學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-106-1.pdf 目前未授權公開取用 | 1.94 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。