目的 : 評估肝細胞癌 (Hepatocellular Carcinoma, HCC)病患接受放射治療 (Radiation Therapy, RT)後誘發肝臟毒性併發症 (Radiation - Induced Hepatic Toxicity, RIHT)之劑量 - 體積參數因子 (Dose - Volume Factors)分析。
材料與方法 : 本研究為回溯性利用收集2014 - 2017年間114位接受使用強度調控放射治療 (Intensity Modulation Radiation Therapy, IMRT)肝細胞癌 (HCC)病患，其中排除肝臟器官接收劑量 - 體積未達到700cc與極端值之病患並利用重複抽樣估計法 (Bootstrapping)增加一倍樣本數達到138筆。根據病患療程期間抽血數值評估上升程度使用常見不良事件評價標準表 (Common Terminology Criteria For Adverse Events, CTCAE)判定有無發生放射治療 (RT)後誘發肝臟毒性併發症 (RIHT)。採用病患正常肝臟接收劑量 - 體積 (Normal Liver Dose - Volume Receiving)和整體肝臟接收劑量 - 體積 (Total Liver Dose - Volume Receiving)以及使用萃取演算法：最小絕對壓縮挑選運算子 (Least Absolute Shrinkage and Selection Operator, LASSO)和貝氏網路因子強度挑選 (Bayesian Network, BN)預測因子，並建立效能演算法模型：邏輯斯迴歸 (Logistic Regression, LR)與單純貝葉斯 (Naive Bayes, NB)最後利用受試者操作特性曲線下面積 (The Area Under the receiver operating characteristic Curve, AUC)、準確度 (Accuracy, ACC)和陰性預測值 (Negative predictive value, NPV)比較模型相關差異性。
結果 : 針對在病患經過放射治療 (RT)誘發肝臟毒性併發症 (RIHT)預測風險劑量 - 體積參數因子在最小絕對壓縮挑選運算子 (LASSO)萃取於邏輯斯迴歸 (LR)建立模型比較結果準確度 (ACC)、受試者操作特性曲線下面積 (AUC)和陰性預測值 (NPV)為正常肝臟接收體積30 Gy (Normal Liver Volume Receiving 30 Gy, NLV30 Gy)，0.76、0.65、0.72與整體肝臟接收體積30 Gy (Total Liver Volume Receiving 30 Gy, TLV30 Gy)，0.84、0.72、0.72，在貝氏網路因子強度挑選 (BN)萃取於單純貝葉斯 (NB)建立模型為整體肝臟接收體積45 & 20 Gy (Total Liver Volume Receiving 45 & 20 Gy, TLV45 & 20 Gy)，0.72、0.76、0.73。
結論 : 本研究結果顯示對於無法切除肝細胞癌 (HCC)患者接受放射治療 (RT)後避免誘發肝臟毒性併發症 (RIHT)之產生應控制在：正常肝臟接受體積V30 Gy (NLV30 Gy)、整體肝臟接受體積V30 Gy (TLV30 Gy)及整體肝臟接受體積V20 Gy (TLV20 Gy)劑量 - 體積參數因子 (Dose - Volume Factors)，並達到肝臟器官對於輻射低耐受度限制與提供醫師做為治療規劃上之參考，及修正合適之治療計畫的指標，希望能藉此達到預防誘發肝臟毒性併發症 (RIHT)之風險。

Purpose : To assess the dose-volume factors analysis of radiation - induced hepatic toxicity (RIHT) in patients with hepatocellular carcinoma (HCC) after radiotherapy.
Materials and methods : The study is retrospectively collected between 2014 and 2017, 114 patients with HCC underwent intensity modulation radiation therapy (IMRT). Of those, liver dose - volume receiving of less than 700 cc and extreme value excluded with remain only 69 patients and to use the bootstrapping of reached 138 samples. Assess blood draw rises level during the course of the patient's treatment RIHT was scored using the Common Terminology Criteria for Adverse Events (CTCAE). The predictor's analysis, we use patient’s dose - volume factors the may cause RIHT, And two kinds of selection algorithm by least absolute shrinkage and selection operator (LASSO), bayesian network (BN) to select the risk predictors. Furthermore, the classification models of the selected risk predictors, all risk predictors were established two kinds of algorithms by logistic regression (LR), naive bayes (NB). Finally, uses the area under the receiver operating characteristic curve (AUC), accuracy (ACC) and negative predictive value (NPV) was compared.
Results : The patients predicted risk dose-volume factors of classification models with AUC, ACC and NPV. First, LASSO select on LR models：normal liver volume receiving 30 Gy (NLV30), 0.76, 0.65, 0.72. Second LASSO select on LR models：total liver volume receiving 30 Gy (TLV30), 0.84, 0.72, 0.72. Third BN select on NB models：total liver volume receiving 45 & 20 Gy (TLV45&20), 0.72, 0.76, 0.73.
Conclusions : The results of this study showed should be limited to the NLV30 Gy, TLV30 Gy, TLV20 Gy as risk predictors of RIHT after IMRT for unresectable HCC, and could reach liver organ with low tolerance for radiation. The advantage of selection factors can be provided to physicians for better decision-making.

摘要 i
Abstract iii
致謝 v
目錄 vi
表目錄 viii
圖目錄 ix
符號縮寫 x
符號說明 xii
第一章 緒論 1
1.1 動機 1
1.2 目的 3
1.3 相關文獻探討 4
第二章 材料與方法 6
2.1 前言 6
2.2 病患資料 7
2.3 併發症評估 8
2.4 放射治療技術 10
2.5 候選因子 10
2.5.1 候選因子定義 11
2.6萃取因子演算法 12
2.6.1 最小絕對壓縮挑選運算子 12
2.6.1.1 最小絕對壓縮挑選運算子挑選之步驟 13
2.6.2 貝氏網路因子強度挑選 14
2.6.2.1 貝氏網路因子強度挑選之步驟 15
2.7 模型效能演算法 16
2.7.1 邏輯斯回歸模型 16
2.7.2 單純貝葉斯模型 17
2.8 模型效能評估 18
2.8.1 準確率 18
2.8.2 受試者操作曲線下面積 18
2.8.3 陰性預測值 18
2.9 正常組織併發症模型 19
第三章 結果 20
3.1 前言 20
3.2 樣本平均劑量圖 21
3.3 樣本因子相關性 22
3.4 因子萃取結果 24
3.4.1 最小絕對壓縮挑選運算子挑選之結果 24
3.4.2 貝氏網路因子強度挑選之結果 26
3.5 因子預測模型 29
3.5.1 邏輯斯回歸模型之效能 29
3.5.2 單純貝葉斯模型之效能 30
3.5.3 模型係數與勝算比 31
3.6 正常組織併發症模型 32
第四章 討論 34
4.1 前言 34
4.2 演算法效能驗證 35
4.2.1 最小絕對壓縮挑選運算子之驗證 35
4.2.2 貝氏網路因子強度挑選之驗證 36
4.2.3 邏輯斯回歸與單純貝葉斯模型之比較 38
4.3 治療計畫指標 40
4.3.1 正常肝臟器官治療指標 41
4.3.2 整體肝臟器官治療指標 44
4.4 正常組織併發症模型建立 47
第五章 結論 50
參考文獻 51
自傳 56

[1]I. A. f. R. on and Cancer, "World Health Organization. Cancer today," 2018.
[2]J. Llovet, M. Ducreux, R. Lencioni, A. Di Bisceglie, P. Galle, and J. Dufour, "European Association for the Study of the Liver European Organisation for Research and Treatment of Cancer: EASL-EORTC clinical practice guidelines: management of hepatocellular carcinoma," J Hepatol, vol. 56, no. 4, pp. 908-943, 2012.
[3]R. Lozano et al., "Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010," The lancet, vol. 380, no. 9859, pp. 2095-2128, 2012.
[4]T. H. Kim et al., "Dose–volumetric parameters predicting radiation-induced hepatic toxicity in unresectable hepatocellular carcinoma patients treated with three-dimensional conformal radiotherapy," International Journal of Radiation Oncology* Biology* Physics, vol. 67, no. 1, pp. 225-231, 2007.
[5]J. C.-H. CHENG, et al., "Radiation-induced liver disease after radiotherapy for hepatocellular carcinoma: clinical manifestation and dosimetric description.," Radiotherapy and oncology, vol. 63, no. 1, pp. 41-45, 2002.
[6]M. Fuss, B. J. Salter, T. S. Herman, and C. R. Thomas Jr, "External beam radiation therapy for hepatocellular carcinoma: potential of intensity-modulated and image-guided radiation therapy," Gastroenterology, vol. 127, no. 5, pp. S206-S217, 2004.
[7]S. H. Son et al., "Stereotactic body radiotherapy for patients with unresectable primary hepatocellular carcinoma: dose-volumetric parameters predicting the hepatic complication," International Journal of Radiation Oncology* Biology* Physics, vol. 78, no. 4, pp. 1073-1080, 2010.
[8]M. Velec et al., "Predictors of liver toxicity following stereotactic body radiation therapy for hepatocellular carcinoma," International Journal of Radiation Oncology* Biology* Physics, vol. 97, no. 5, pp. 939-946, 2017.
[9]e. a. INTENSITY MODULATED RADIATION THERAPY COLLABORATIVE WORKING GROUP, " Intensity-modulated radiotherapy: current status and issues of interest. ," International Journal of Radiation Oncology* Biology* Physics, vol. 51, no. 4, pp. 880-914, 2001.
[10]S. I. GUTIONTOV, et al., "Intensity‐modulated radiotherapy for head and neck surgeons.," Head & neck, vol. 38, no. S1, pp. E2368-E2373, 2016.
[11]S. K. TANGUTURI, et al., "Radiation therapy for liver tumors: ready for inclusion in guidelines?." The oncologist, vol. 19, no. 8, pp. 868-879, 2014.
[12]M. FUSS, et al., "External beam radiation therapy for hepatocellular carcinoma: potential of intensity-modulated and image-guided radiation therapy. ," Gastroenterology, vol. 127, no. 5, pp. S206-S217., 2004.
[13]C. H. S. RIM, Jinsil., "Application of radiotherapy for hepatocellular carcinoma in current clinical practice guidelines. ," Radiation oncology journal, vol. 34, no. 3, p. 160, 2016.
[14]F. K. KEANE, et al., "Liver-directed radiotherapy for hepatocellular carcinoma.," Liver Cancer, vol. 5, no. 3, pp. 198-209, 2016.
[15]L. A. Dawson and R. K. Ten Haken, "Partial volume tolerance of the liver to radiation," in Seminars in radiation oncology, 2005, vol. 15, no. 4, pp. 279-283: Elsevier.
[16]J. C.-H. Cheng et al., "Radiation-induced liver disease after three-dimensional conformal radiotherapy for patients with hepatocellular carcinoma: dosimetric analysis and implication," International Journal of Radiation Oncology* Biology* Physics, vol. 54, no. 1, pp. 156-162, 2002.
[17]J. C.-H. Cheng et al., "Biologic susceptibility of hepatocellular carcinoma patients treated with radiotherapy to radiation-induced liver disease," International Journal of Radiation Oncology* Biology* Physics, vol. 60, no. 5, pp. 1502-1509, 2004.
[18]S.-X. Liang et al., "Radiation-induced liver disease in three-dimensional conformal radiation therapy for primary liver carcinoma: the risk factors and hepatic radiation tolerance," International Journal of Radiation Oncology* Biology* Physics, vol. 65, no. 2, pp. 426-434, 2006.
[19]I. K. Yamada K, Sugimoto K, et al., "Prospective trial of combined transcatheter arterial chemoembolization and threedimensional conformal radiotherapy for portal vein tumor thrombus in patients with unresectable hepatocellular carcinoma.," Int J Radiat Oncol Biol Phys, vol. 2003, no. 57, pp. 113-119, 2003.
[20]W. J. Cheng JC, Huang CM, et al., "Radiation-induced liver disease after three-dimensional conformal radiotherapy for patients with hepatocellular carcinoma: Dosimetric analysis and implication.," Int J Radiat Oncol Biol Phys, vol. 2002, no. 54, pp. 156-162, 2002.
[21]J. Ingold, "Radiation hepatitis," AJR, vol. 93, pp. 200-208, 1965.
[22]K. Lewin and R. Millis, "Human radiation hepatitis. A morphologic study with emphasis on the late changes," Archives of pathology, vol. 96, no. 1, pp. 21-26, 1973.
[23]B. C. Kutcher GJ, "Calculation of complication probability factors for non-uniform normal tissue irradiation: The effective volume method.," Int J Radiat Oncol Biol Phys, vol. 16, no. 1, pp. 1623-1630, 1989.
[24]C. Burman, G. Kutcher, B. Emami, and M. Goitein, "Fitting of normal tissue tolerance data to an analytic function," International Journal of Radiation Oncology* Biology* Physics, vol. 21, no. 1, pp. 123-135, 1991.
[25]T. S. Lawrence et al., "The use of 3-D dose volume analysis to predict radiation hepatitis," International Journal of Radiation Oncology* Biology* Physics, vol. 23, no. 4, pp. 781-788, 1992.
[26]N. D. Dawson LA, Balter JM, et al., "Analysis of radiation-induced liver disease using the Lyman NTCP model.," Int J Radiat Oncol Biol Phys, vol. 53, pp. 810-821, 2002.
[27]S. P. Lee, M. Y. Leu, J. B. Smathers, W. H. McBride, R. G. Parker, and H. R. Withers, "Biologically effective dose distribution based on the linear quadratic model and its clinical relevance," International Journal of Radiation Oncology* Biology* Physics, vol. 33, no. 2, pp. 375-389, 1995.
[28]R. Tibshirani, "Regression shrinkage and selection via the lasso," Journal of the Royal Statistical Society: Series B (Methodological), vol. 58, no. 1, pp. 267-288, 1996.
[29]H. Zou, "The adaptive lasso and its oracle properties," Journal of the American statistical association, vol. 101, no. 476, pp. 1418-1429, 2006.
[30]J. Pearl, "Bayesian networks," 2011.
[31]N. Friedman, D. Geiger, and M. Goldszmidt, "Bayesian network classifiers," Machine learning, vol. 29, no. 2-3, pp. 131-163, 1997.
[32]維基百科, "貝氏網路."
[33]S. Menard, Applied logistic regression analysis. Sage, 2002.
[34]A. P. Worth and M. T. Cronin, "The use of discriminant analysis, logistic regression and classification tree analysis in the development of classification models for human health effects," Journal of Molecular Structure: THEOCHEM, vol. 622, no. 1-2, pp. 97-111, 2003.
[35]T. R. Patil and S. Sherekar, "Performance analysis of Naive Bayes and J48 classification algorithm for data classification," International journal of computer science and applications, vol. 6, no. 2, pp. 256-261, 2013.
[36]G. Dimitoglou, J. A. Adams, and C. M. Jim, "Comparison of the C4. 5 and a Naïve Bayes classifier for the prediction of lung cancer survivability," arXiv preprint arXiv:1206.1121, 2012.
[37]M. H. Dunham, Data mining: Introductory and advanced topics. Pearson Education India, 2006.
[38]S. Youn and D. McLeod, "A comparative study for email classification," in Advances and innovations in systems, computing sciences and software engineering: Springer, 2007, pp. 387-391.
[39]B. Wang, "Variable ranking by solution-path algorithms," University of Waterloo, 2012.
[40]S. H. Park, J. M. Goo, and C.-H. Jo, "Receiver operating characteristic (ROC) curve: practical review for radiologists," Korean Journal of Radiology, vol. 5, no. 1, pp. 11-18, 2004.
[41]T.-F. Lee and F.-M. Fang, "Quantitative analysis of normal tissue effects in the clinic (QUANTEC) guideline validation using quality of life questionnaire datasets for parotid gland constraints to avoid causing xerostomia during head-and-neck radiotherapy," Radiotherapy and Oncology, vol. 106, no. 3, pp. 352-358, 2013.
[42]J. T. Lyman, "Complication probability as assessed from dose-volume histograms," Radiation Research, vol. 104, no. 2s, pp. S13-S19, 1985.
[43]M.-H. LIOU, "Predictive Performances Analysis of Statistical Learning Methods on Retrospective Multivariate Normal Tissue Complication Probability Models Investigating Causing Xerostomia after Head and Neck Cancer Radiotherapy," 2016.
[44]S. H. Bae, W. I. Jang, and H. C. Park, "Intensity-modulated radiotherapy for hepatocellular carcinoma: dosimetric and clinical results," Oncotarget, vol. 8, no. 35, p. 59965, 2017.
[45]Y.-C. Kuo et al., "Volumetric intensity-modulated Arc (RapidArc) therapy for primary hepatocellular carcinoma: comparison with intensity-modulated radiotherapy and 3-D conformal radiotherapy," Radiation Oncology, vol. 6, no. 1, p. 76, 2011.
[46]T. S. Lawrence, J. M. Robertson, M. S. Anscher, R. L. Jirtle, W. D. Ensminger, and L. F. Fajardo, "Hepatic toxicity resulting from cancer treatment," International Journal of Radiation Oncology* Biology* Physics, vol. 31, no. 5, pp. 1237-1248, 1995.
[47]T.-F. Lee et al., "Using multivariate regression model with least absolute shrinkage and selection operator (LASSO) to predict the incidence of xerostomia after intensity-modulated radiotherapy for head and neck cancer," PloS one, vol. 9, no. 2, p. e89700, 2014.
[48]M. Acalovschi et al., "Risk factors for symptomatic gallstones in patients with liver cirrhosis: a case–control study," The American journal of gastroenterology, vol. 98, no. 8, pp. 1856-1860, 2003.
[49]M. L. Wong, W. Lam, K. S. Leung, P. S. Ngan, and J. C. Cheng, "Discovering knowledge from medical databases using evolutionory algorithms," IEEE Engineering in Medicine and Biology Magazine, vol. 19, no. 4, pp. 45-55, 2000.
[50]H. Jin, S. Kim, and J. Kim, "Decision factors on effective liver patient data prediction," International journal of Bio-science and Bio-Technology, vol. 6, no. 4, pp. 167-178, 2014.
[51]B. V. Ramana, M. S. P. Babu, and N. Venkateswarlu, "A critical study of selected classification algorithms for liver disease diagnosis," International Journal of Database Management Systems, vol. 3, no. 2, pp. 101-114, 2011.
[52]K. Okuda et al., "Natural history of hepatocellular carcinoma and prognosis in relation to treatment study of 850 patients," Cancer, vol. 56, no. 4, pp. 918-928, 1985.
[53]J. M. Robertson, T. S. Lawrence, J. C. Andrews, S. Walker, M. L. Kessler, and W. D. Ensminger, "Long-term results of hepatic artery fluorodeoxyuridine and conformal radiation therapy for primary hepatobiliary cancers," International Journal of Radiation Oncology* Biology* Physics, vol. 37, no. 2, pp. 325-330, 1997.
[54]C. o. t. L. I. P. Investigators, "A new prognostic system for hepatocellular carcinoma: a retrospective study of 435 patients," Hepatology, vol. 28, no. 3, pp. 751-755, 1998.
[55]J. Jung et al., "Radiation-induced liver disease after stereotactic body radiotherapy for small hepatocellular carcinoma: clinical and dose-volumetric parameters," Radiation oncology, vol. 8, no. 1, p. 249, 2013.
[56]P. Liang et al., "Effect of CyberKnife stereotactic body radiation therapy for hepatocellular carcinoma on hepatic toxicity," OncoTargets and therapy, vol. 9, p. 7169, 2016.
[57]S.-X. Liang et al., "Dosimetric predictor identification for radiation-induced liver disease after hypofractionated conformal radiotherapy for primary liver carcinoma patients with Child–Pugh Grade A cirrhosis," Radiotherapy and Oncology, vol. 98, no. 2, pp. 265-269, 2011.