高雄醫學大學圖書館 |

Language: English

Back

結直腸癌患者在高血糖環境中放射線治療合併化學藥物療效之放射組學關聯性 =...

楊舒涵

結直腸癌患者在高血糖環境中放射線治療合併化學藥物療效之放射組學關聯性 =Radiomic correlation of the efficacy of combined radiation therapy and chemotherapy in patients with colorectal cancer in a hyperglycemic environment /

Record Type:	Language materials, printed : Monograph/item
Title/Author:	結直腸癌患者在高血糖環境中放射線治療合併化學藥物療效之放射組學關聯性 =/ 楊舒涵
Reminder of title:	Radiomic correlation of the efficacy of combined radiation therapy and chemotherapy in patients with colorectal cancer in a hyperglycemic environment /
remainder title:	Radiomic correlation of the efficacy of combined radiation therapy and chemotherapy in patients with colorectal cancer in a hyperglycemic environment
Author:	楊舒涵
Published:	[高雄市]: [撰者], : 2025[民114],
Description:	84葉 :圖 ; : 30公分;
Notes:	指導教授: 柯建志, 黃旼儀.
基督教聖經之智慧書導讀 :	背景與目的：結直腸癌在全球以及台灣具高發生率與高死亡率，新輔助同步化學放射治療（neoadjuvant concurrent chemoradiotherapy, NACRT）為二、三期患者常見且有效的策略，然而其中對治療反應不佳的患者，其中不少合併高血糖的背景，高血糖已被證實減弱化學和放射治療效果，增加後續手術的困難，患者須承受無效治療的副作用，也造成整體存活率下降，臨床上目前並無有效方法可精準預測結直腸癌患者NACRT反應的工具，本研究旨在運用放射組學（Radiomics）技術預測結直腸癌患者對NACRT反應，並探討其與高血糖背景的關聯性。材料與方法：本研究經人體試驗委員會同意後，回溯性收集92名接受NACRT結直腸癌患者的臨床資訊，擷取其術前放射治療前定位的電腦斷層影像，使用LIFEx放射組學軟體萃取腫瘤放射組學特徵，隨機將患者分為訓練組（n = 65）與驗證組（n = 27），以LASSO邏輯斯回歸於訓練組中篩選具完全病理反應（pathological complete response, pCR）預測力之特徵，建立模型並以ROC曲線評估其效能。同時將放射組學分數與臨床變項（如CEA、HbA1c等）進行單變量與多變量邏輯斯回歸分析，建立pCR預測模型。結果：自LIFEx擷取190項放射組學特徵中，111項具可用性，經訓練組（n = 65）中進行LASSO迴歸分析，選7項特徵建立預測模型，此模型在訓練組和驗證組中預測pCR和Non-pCR患者均有良好準確度（AUC = 0.854和0.818），多變量分析顯示CEA和放射組學分數為pCR顯著預測因子，與血糖相關的因子（HbA1c與空腹血糖值）則未顯示與pCR具有統計顯著關聯，最終預測模型經500次Bootstrap Resampling驗證，顯示良好校準（平均誤差分別爲0.042和0.045）。結論：本研究所建構之整合放射組學特徵與CEA的預測模型可有效預測結直腸癌患者於NACRT後是否達成pCR，具良好準確度與穩定性，對於個體化治療策略之發展具有潛在應用價值。.
Online resource:	電子資源
[NT 15000934]:	(平裝)

結直腸癌患者在高血糖環境中放射線治療合併化學藥物療效之放射組學關聯性 =Radiomic correlation of the efficacy of combined radiation therapy and chemotherapy in patients with colorectal cancer in a hyperglycemic environment /
楊舒涵

結直腸癌患者在高血糖環境中放射線治療合併化學藥物療效之放射組學關聯性 =Radiomic correlation of the efficacy of combined radiation therapy and chemotherapy in patients with colorectal cancer in a hyperglycemic environment /Radiomic correlation of the efficacy of combined radiation therapy and chemotherapy in patients with colorectal cancer in a hyperglycemic environment楊舒涵 - [高雄市]: [撰者], 2025[民114] - 84葉 :圖 ;30公分

指導教授: 柯建志, 黃旼儀.

碩士論文--高雄醫學大學醫學影像暨放射科學系碩士在職專班.

參考書目：葉.

第一章緒論1

背景與目的：結直腸癌在全球以及台灣具高發生率與高死亡率，新輔助同步化學放射治療（neoadjuvant concurrent chemoradiotherapy, NACRT）為二、三期患者常見且有效的策略，然而其中對治療反應不佳的患者，其中不少合併高血糖的背景，高血糖已被證實減弱化學和放射治療效果，增加後續手術的困難，患者須承受無效治療的副作用，也造成整體存活率下降，臨床上目前並無有效方法可精準預測結直腸癌患者NACRT反應的工具，本研究旨在運用放射組學（Radiomics）技術預測結直腸癌患者對NACRT反應，並探討其與高血糖背景的關聯性。材料與方法：本研究經人體試驗委員會同意後，回溯性收集92名接受NACRT結直腸癌患者的臨床資訊，擷取其術前放射治療前定位的電腦斷層影像，使用LIFEx放射組學軟體萃取腫瘤放射組學特徵，隨機將患者分為訓練組（n = 65）與驗證組（n = 27），以LASSO邏輯斯回歸於訓練組中篩選具完全病理反應（pathological complete response, pCR）預測力之特徵，建立模型並以ROC曲線評估其效能。同時將放射組學分數與臨床變項（如CEA、HbA1c等）進行單變量與多變量邏輯斯回歸分析，建立pCR預測模型。結果：自LIFEx擷取190項放射組學特徵中，111項具可用性，經訓練組（n = 65）中進行LASSO迴歸分析，選7項特徵建立預測模型，此模型在訓練組和驗證組中預測pCR和Non-pCR患者均有良好準確度（AUC = 0.854和0.818），多變量分析顯示CEA和放射組學分數為pCR顯著預測因子，與血糖相關的因子（HbA1c與空腹血糖值）則未顯示與pCR具有統計顯著關聯，最終預測模型經500次Bootstrap Resampling驗證，顯示良好校準（平均誤差分別爲0.042和0.045）。結論：本研究所建構之整合放射組學特徵與CEA的預測模型可有效預測結直腸癌患者於NACRT後是否達成pCR，具良好準確度與穩定性，對於個體化治療策略之發展具有潛在應用價值。.

(平裝)Subjects--Index Terms:

糖化血色素

結直腸癌患者在高血糖環境中放射線治療合併化學藥物療效之放射組學關聯性 =Radiomic correlation of the efficacy of combined radiation therapy and chemotherapy in patients with colorectal cancer in a hyperglycemic environment /
LDR:05694cam a2200229 i 4500 001 386760
008 250923s2025 ch a e m 000 0 chi d
035 $a (THES)113KMC0077E007
100 1 $a 楊舒涵 $3 512834
245 1 0 $a 結直腸癌患者在高血糖環境中放射線治療合併化學藥物療效之放射組學關聯性 = $b Radiomic correlation of the efficacy of combined radiation therapy and chemotherapy in patients with colorectal cancer in a hyperglycemic environment / $c 楊舒涵
246 1 1 $a Radiomic correlation of the efficacy of combined radiation therapy and chemotherapy in patients with colorectal cancer in a hyperglycemic environment
260 # $a [高雄市]: $b [撰者], $c 2025[民114]
300 $a 84葉 : $b 圖 ; $c 30公分
500 $a 指導教授: 柯建志, 黃旼儀.
500 $a 學年度: 113.
502 $a 碩士論文--高雄醫學大學醫學影像暨放射科學系碩士在職專班.
504 $a 參考書目：葉.
505 0 # $a 第一章緒論1
520 3 $a 背景與目的：結直腸癌在全球以及台灣具高發生率與高死亡率，新輔助同步化學放射治療（neoadjuvant concurrent chemoradiotherapy, NACRT）為二、三期患者常見且有效的策略，然而其中對治療反應不佳的患者，其中不少合併高血糖的背景，高血糖已被證實減弱化學和放射治療效果，增加後續手術的困難，患者須承受無效治療的副作用，也造成整體存活率下降，臨床上目前並無有效方法可精準預測結直腸癌患者NACRT反應的工具，本研究旨在運用放射組學（Radiomics）技術預測結直腸癌患者對NACRT反應，並探討其與高血糖背景的關聯性。材料與方法：本研究經人體試驗委員會同意後，回溯性收集92名接受NACRT結直腸癌患者的臨床資訊，擷取其術前放射治療前定位的電腦斷層影像，使用LIFEx放射組學軟體萃取腫瘤放射組學特徵，隨機將患者分為訓練組（n = 65）與驗證組（n = 27），以LASSO邏輯斯回歸於訓練組中篩選具完全病理反應（pathological complete response, pCR）預測力之特徵，建立模型並以ROC曲線評估其效能。同時將放射組學分數與臨床變項（如CEA、HbA1c等）進行單變量與多變量邏輯斯回歸分析，建立pCR預測模型。結果：自LIFEx擷取190項放射組學特徵中，111項具可用性，經訓練組（n = 65）中進行LASSO迴歸分析，選7項特徵建立預測模型，此模型在訓練組和驗證組中預測pCR和Non-pCR患者均有良好準確度（AUC = 0.854和0.818），多變量分析顯示CEA和放射組學分數為pCR顯著預測因子，與血糖相關的因子（HbA1c與空腹血糖值）則未顯示與pCR具有統計顯著關聯，最終預測模型經500次Bootstrap Resampling驗證，顯示良好校準（平均誤差分別爲0.042和0.045）。結論：本研究所建構之整合放射組學特徵與CEA的預測模型可有效預測結直腸癌患者於NACRT後是否達成pCR，具良好準確度與穩定性，對於個體化治療策略之發展具有潛在應用價值。.
520 3 $a Background and objective: Colorectal cancer (CRC) shows high incidence and mortality worldwide, including in Taiwan. Neoadjuvant concurrent chemoradiotherapy (NACRT) is an effective strategy for stage II–III patients. However, a subset of patients demonstrates poor therapeutic response, among whom a significant proportion presents with hyperglycemia. Hyperglycemia has been shown to impair the efficacy of both chemotherapy and radiotherapy, complicating subsequent surgical procedures and exposing patients to the adverse effects of ineffective treatments, thereby compromising overall survival. Currently, no reliable clinical tool is available to accurately predict individual NACRT response in CRC patients. This study aims to develop a radiomics-based model to predict NACRT response in CRC patients and to investigate its correlation with hyperglycemic status. Materials and methods: After approval by the institutional review board, this retrospective study enrolled 92 patients with mid-to-late stage colorectal cancer who underwent NACRT. Pre-treatment planning CT images were collected, and radiomic features were extracted using the open-source software LIFEx. Patients were randomly divided into a training cohort (n = 65) and a validation cohort (n = 27). Least absolute shrinkage and selection operator (LASSO) logistic regression was performed in the training cohort to select features predictive of pathological complete response (pCR).A prediction model was constructed and its performance was evaluated using the area under the receiver operating characteristic curve (AUC). Results: Among the 190 radiomic features initially extracted using LIFEx, 111 were determined to be usable. Seven key features were selected by LASSO regression in the training cohort and used to construct the predictive model. The model demonstrated good discriminative ability in both the training and validation cohorts for identifying pCR and non-pCR patients (AUC = 0.854 and 0.818, respectively). Conclusion: The prediction model developed in this study, integrating radiomic features and CEA levels, demonstrated high accuracy and robustness in predicting pCR following NACRT in colorectal cancer patients. This model shows potential value in guiding individualized treatment strategies and improving clinical decision-making in the context of precision oncology..
563 $a (平裝)
653 # # $a 糖化血色素 $a 結直腸癌 $a 放射組學 $a 電腦斷層影像.
653 # # $a HbA1c $a colorectal cancer $a radiomics $a computed tomography imaging.
856 7 # $u https://handle.ncl.edu.tw/11296/wb8qxb $z 電子資源 $2 http