- 學位論文
探討18F-FDG PET SUVmax與生理參數之關聯性
Investigating the Relationship between SUVmax of 18F-FDG PET and Physiological Parameters
摘要
正子發射斷層攝影(Positron Emission Tomography, PET)造影技術，是目前臨床對於腫瘤的偵測、分期及治療效果的評估與追蹤使用最廣泛的診斷工具。臨床醫師除了利用肉眼判讀PET影像作為主要的診斷依據，其最大標準攝取值(Maximum Standard Uptake Value, SUVmax)亦能提供PET影像像素的量化數據，輔佐醫師診斷並提高診斷的準確度。本研究將利用年齡進行分組，探討其大腦、小腦、心肌壁、肝臟及肺臟的最大標準攝取值與半高全寬(Full Width at Half Maximum, FWHM)、年齡、身體質量指數(Body Mass Index, BMI)及體重之關聯性。 本研究採用回溯性分組實驗設計，收集經核醫專科醫師診斷之無異常報告及18F-FDG正子電腦斷層影像(18F-FDG PET/CT)共90筆，並使用2D手動圈選法及區域成長法計算大腦、小腦、心肌壁、肝臟及肺臟的SUVmax。根據不同年齡對90筆18F-FDG PET/CT影像進行分組，使用相關性分析探討大腦、小腦、心肌壁、肝臟及肺臟的SUVmax與FWHM、年齡、BMI、體重的相關性。接著，將具顯著相關性的SUVmax與參數，繪製誤差圖、散佈圖及進行線性迴歸分析等方法以建立合理相關性模型提供臨床參考之用途。 結果顯示年齡與大腦、小腦、肝臟、肺臟之SUVmax及BMI與心肌壁SUVmax的相關係數均高達0.90，相關程度為高度相關且線性迴歸模型的模型解釋能力(R2)均高於80%，隨着年齡改變將會影響大腦、小腦、肝臟及肺臟的葡萄糖攝取量，另外隨著BMI改變會影響心肌壁的葡萄糖攝取量。因此本研究的相關性模型可作為成人健康檢查無異常疾病健康族群葡萄糖代謝率的參考。未來仍需增加無異常者的樣本數及異常的案例一併納入討論，以瞭解相關性模型的變化情況。
並列摘要
Positron Emission Tomography (PET) imaging is widely used in clinical diagnosis for tumor detection, staging and treatment evaluation. In addition to the clinician interpret PET images with the naked eye as the main basis for the diagnosis, the maximum standard uptake value (SUVmax) can also provide quantitative data from PET imaging and improve the accuracy of diagnosis. This study was used conducted age group to investigate the correlation of SUVmax from brain, cerebellum, myocardial wall, liver and lungs with Full Width at Half Maximum (FWHM), Body Mass Index (BMI) and body weight. This study is a retrospective experiment, collected totally 90 cases of normal 18F-FDG PET/CT images by nuclear medicine physician diagnostic report. The 2D manually draw method and seed region growing method were used to calculate the SUVmax from brain, cerebellum, heart, liver and lungs. According to the different age group of 90 cases explored the relation among the SUVmax from brain, cerebellum, myocardial wall, liver and lungs with FWHM, age, BMI and body weight by using correlation analysis. Subsequently, the significant correlation with the parameters of SUVmax were be used to draw the error plot, scatter plot. The linear regression analysis was used to establish correlated models to providing clinical assistant reference purposes. The relation coefficient between age and brain, cerebellum, liver, lungs SUVmax were higher than 0.90 in this study. The R2 of linear regression models between age and brain, cerebellum, liver, lungs SUVmax were higher than 80%. The SUVmax of brain, cerebellum, liver and lungs are varieties with patient’s age. Meanwhile, the glucose uptake of myocardial wall is affected by the index of BMI. Hence, the correlated models in this study might be used to a reference of glucose metabolism rate for healthy adult population. There are more abnormal and normal cases involved in the future study in order to figure out the details variety of glucose metabolism rates.