鍾孝文臺灣大學：電機工程學研究所劉華姍Liu, Hua-ShanHua-ShanLiu2007-11-262018-07-062007-11-262018-07-062004http://ntur.lib.ntu.edu.tw//handle/246246/53383在臨床放射線治療計劃的設計上，一個可靠而且安全的治療計劃是希望能夠達到將放射線的高劑量區集中在腫瘤部分,同時希望避開腫瘤附近的正常組織，使正常組織的吸收劑量降到最低，以達到除去腫瘤而且兼顧保護正常細胞弁鄋熙戽蚰堛滿C因為這個重要的理由，我們期待藉由這個研究，代入現有病人的劑量體積資料，來找出最適合的參數組合，建立一個放射線生物機率的模型，將輻射生物效應加以定量之。在這個研究中，我們引入的是由平行性器官特性所導證出的平行性器官的正常組織併發症機率模型(The parallel architecture NTCP model)，這個機率模型的基本假設是建立在一個照射體積閥值的觀念上，如果一個正常組織接受放射線照射後，所受到的傷害體積經過放射線劑量的加權超過既定的閥值，那麼這個正常器官的弁鉥N會受到影響，反之，如果正常組織受到傷害的體積沒有超過閥值，那麼整個正常器官的弁鉥N得以保存下來。這個研究主要放入四組不同的正常肝細胞的劑量體積資料，這四組體積資料是由兩種不同病理特徵(肝癌和胃癌)的病人群體，再利用有沒有B型肝炎的帶原再分成共四小組子群體。利用統計方法中的最大可能性法(maximum likelihood method)，我們分別找到了4組最佳的參數組合，代入傷害體積的計算公式，發現四組中有併發症的病人其計算出的傷害體積值皆在0.4以上。但是只有未帶原B型肝炎的2組病人群體，其有無併發症的傷害體積值才有明顯的差別，也就是說，如果不把人數的限制列入考量的範圍，只有這兩組病人群體才能比較合理的以此種機率模型解釋之。An optimized treatment plan procedure of radiation therapy (RT) is done by delivering a sufficient dose of radiation to eradicate cancerous cells without having severe complication in healthy organs. For this reason, radiobiological models should be introduced to quantify the radiobiological response of normal tissues to radiotherapy. To predict the incidence of radiation-induced disease in parallel organs, the parallel architecture normal tissue complication probability (NTCP) model hypothesizes that a complication would be caused if the damaged fraction of the organ volume (f) exceeds the threshold of a critical one, ft (the “ functional reserve”). This study analyzed data of normal liver function from 4 patient subgroups with the parallel NTCP model to investigate the tolerance of the partial liver irradiation by evaluating the value of f. The calculated f from maximum likelihood estimated (MLE) parameters shows that cases of radiation-induced liver disease (RILD) have been reported in patients with a value of f larger than 0.4.English Abstract Chinese Abstract 1. INTRODUCTION….………………………………………………….5 2. THREE-DIMENSIONAL CONFORMAL RADIOTHERAPY (3DCRT)……………………………………………………………….7 2-1. Localization and immobilization of a treatment position for the patient……………………………………………………..9 2-2. CT imaging……………………………………………………….10 2-3. Delineation of target volume……………………………………..11 2-4. Beam designing and dose calculation…………………………….12 2-5. Plan optimization and evaluation…………………………………13 3. MATERIALS AND METHODS……………………………………....33 3-1.Patients…………………………………………………………….33 3-2. Parallel Architecture NTCP Model……………………………….34 3-3. The Method of Maximum Likelihood……………...…………….38 3-4. Goodness of fit……………………………………………………41 4. RESULT………………………………………………………………49 4.1. MLE parameters, confidence intervals and goodness of fit……...49 4.2 MLE NTCP and calculated damaged fractional volume………….51 4.3 Observed vs. predicted complication rate…………………………52 4.4 Threshold of damaged fractional volume and irradiated dose……53 5. DISCUSSION………………………..………………………………..67 6. CONCLUSION………………………………………………………..73 APPENDIX REFERENCES………………………………………………………..….753792975 bytesapplication/pdfen-US生物機率模型平行性器官parallel architectureNTCP model[SDGs]SDG3thesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/53383/1/ntu-93-R91921112-1.pdf