Skip navigation

DSpace

機構典藏 DSpace 系統致力於保存各式數位資料(如:文字、圖片、PDF)並使其易於取用。

點此認識 DSpace
DSpace logo
English
中文
  • 瀏覽論文
    • 校院系所
    • 出版年
    • 作者
    • 標題
    • 關鍵字
    • 指導教授
  • 搜尋 TDR
  • 授權 Q&A
    • 我的頁面
    • 接受 E-mail 通知
    • 編輯個人資料
  1. NTU Theses and Dissertations Repository
  2. 工學院
  3. 醫學工程學研究所
請用此 Handle URI 來引用此文件: http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/81948
完整後設資料紀錄
DC 欄位值語言
dc.contributor.advisor陳中明(Chung-Ming Chen)
dc.contributor.authorChuan-Wei Wangen
dc.contributor.author汪傳崴zh_TW
dc.date.accessioned2022-11-25T03:07:14Z-
dc.date.available2027-02-08
dc.date.copyright2022-02-18
dc.date.issued2022
dc.date.submitted2022-02-09
dc.identifier.citation[1] 108年國人死因統計結果-衛生福利部。民國108年06月16日。檢自https://www.mohw.gov.tw/cp-4631-54482-1.html [2] Siegel, Rebecca L., Kimberly D. Miller, and Ahmedin Jemal. 'Cancer statistics, 2019.' CA: a cancer journal for clinicians 69.1 (2019): 7-34. [3] Detterbeck, Frank C., et al. 'The IASLC Lung Cancer Staging Project: methodology and validation used in the development of proposals for revision of the stage classification of NSCLC in the forthcoming (eighth) edition of the TNM classification of lung cancer.' Journal of thoracic oncology 11.9 (2016): 1433-1446. [4] Goldstraw, Peter, et al. 'The IASLC lung cancer staging project: proposals for revision of the TNM stage groupings in the forthcoming (eighth) edition of the TNM classification for lung cancer.' Journal of Thoracic Oncology 11.1 (2016): 39-51. [5] Brindle, Lucy, et al. 'Eliciting symptoms interpreted as normal by patients with early-stage lung cancer: could GP elicitation of normalised symptoms reduce delay in diagnosis? Cross-sectional interview study.' BMJ open 2.6 (2012). [6] Wood, Douglas E., et al. 'Lung cancer screening.' Journal of the National Comprehensive Cancer Network 10.2 (2012): 240-265. [7] Bach, Peter B., et al. 'Benefits and harms of CT screening for lung cancer: a systematic review.' Jama 307.22 (2012): 2418-2429. [8] Okada, Morihito et al. “Radical sublobar resection for small-sized non-small cell lung cancer: a multicenter study.” The Journal of thoracic and cardiovascular surgery vol. 132,4 (2006): 769-75. [9] El-Sherif, Amgad et al. “Outcomes of sublobar resection versus lobectomy for stage I non-small cell lung cancer: a 13-year analysis.” The Annals of thoracic surgery vol. 82,2 (2006): 408-15; discussion 415-6. [10] Zhong, Chenxi et al. “Comparison of thoracoscopic segmentectomy and thoracoscopic lobectomy for small-sized stage IA lung cancer.” The Annals of thoracic surgery vol. 94,2 (2012): 362-7. [11] Tsutani, Yasuhiro et al. “Oncologic outcomes of segmentectomy compared with lobectomy for clinical stage IA lung adenocarcinoma: propensity score-matched analysis in a multicenter study.” The Journal of thoracic and cardiovascular surgery vol. 146,2 (2013): 358-64. [12] Smith, Cardinale B., et al. 'Survival after segmentectomy and wedge resection in stage I non–small-cell lung cancer.' Journal of Thoracic Oncology 8.1 (2013): 73-78. [13] Cao, Jinlin, et al. 'Survival rates after lobectomy, segmentectomy, and wedge resection for non-small cell lung cancer.' The Annals of thoracic surgery 105.5 (2018): 1483-1491. [14] Landreneau, Rodney J., et al. 'Wedge resection versus lobectomy for stage I (T1 N0 M0) non-small-cell lung cancer.' The Journal of thoracic and cardiovascular surgery 113.4 (1997): 691-700. [15] Travis, William D., et al. 'Visceral pleural invasion: pathologic criteria and use of elastic stains: proposal for the 7th edition of the TNM classification for lung cancer.' Journal of Thoracic Oncology 3.12 (2008): 1384-1390. [16] Kudo, Yujin, et al. 'Impact of visceral pleural invasion on the survival of patients with non-small cell lung cancer.' Lung cancer 78.2 (2012): 153-160. [17] Shimizu, Kimihiro, et al. 'Visceral pleural invasion is an invasive and aggressive indicator of non-small cell lung cancer.' The Journal of thoracic and cardiovascular surgery 130.1 (2005): 160-165. [18] Shimizu, Kimihiro, et al. 'Visceral pleural invasion classification in non–small cell lung cancer: a proposal on the basis of outcome assessment.' The Journal of thoracic and cardiovascular surgery 127.6 (2004): 1574-1578. [19] Koike, Terumoto, et al. 'Risk factor analysis of locoregional recurrence after sublobar resection in patients with clinical stage IA non–small cell lung cancer.' The Journal of thoracic and cardiovascular surgery 146.2 (2013): 372-378. [20] Cao, Christopher, et al. 'Could less be more?—A systematic review and meta-analysis of sublobar resections versus lobectomy for non-small cell lung cancer according to patient selection.' Lung Cancer 89.2 (2015): 121-132. [21] Taioli, Emanuela, et al. 'Survival after sublobar resection for early-stage lung cancer: methodological obstacles in comparing the efficacy to lobectomy.' Journal of Thoracic Oncology 11.3 (2016): 400-406. [22] Jeon, Hyun Woo, et al. 'Sublobar resection versus lobectomy in solid-type, clinical stage IA, non-small cell lung cancer.' World journal of surgical oncology 12.1 (2014): 1-6. [23] Nakagawa, Kazuo et al. “The Lung Cancer Surgical Study Group of the Japan Clinical Oncology Group: past activities, current status and future direction.” Japanese journal of clinical oncology vol. 47,3 (2017): 194-199. [24] El-Baz, Ayman, et al. 'Computer-aided diagnosis systems for lung cancer: challenges and methodologies.' International journal of biomedical imaging 2013 (2013). [25] Doi, Kunio. 'Computer-aided diagnosis in medical imaging: historical review, current status and future potential.' Computerized medical imaging and graphics 31.4-5 (2007): 198-211. [26] Jacobs, Colin, et al. 'Solid, part-solid, or non-solid?: classification of pulmonary nodules in low-dose chest computed tomography by a computer-aided diagnosis system.' Investigative radiology 50.3 (2015): 168-173. [27] Chugh, Gunjan, Shailender Kumar, and Nanhay Singh. 'Survey on Machine Learning and Deep Learning Applications in Breast Cancer Diagnosis.' Cognitive Computation (2021): 1-20. [28] Gillies, Robert J., Paul E. Kinahan, and Hedvig Hricak. 'Radiomics: images are more than pictures, they are data.' Radiology 278.2 (2016): 563-577. [29] Zhao, Xinzhuo, et al. 'Agile convolutional neural network for pulmonary nodule classification using CT images.' International journal of computer assisted radiology and surgery 13.4 (2018): 585-595. [30] LeCun, Yann, et al. 'Gradient-based learning applied to document recognition.' Proceedings of the IEEE 86.11 (1998): 2278-2324. [31] Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. 'Imagenet classification with deep convolutional neural networks.' Advances in neural information processing systems 25 (2012): 1097-1105. [32] Simonyan, Karen, and Andrew Zisserman. 'Very deep convolutional networks for large-scale image recognition.' arXiv preprint arXiv:1409.1556 (2014). [33] Szegedy, Christian, et al. 'Going deeper with convolutions.' Proceedings of the IEEE conference on computer vision and pattern recognition. 2015. [34] He, Kaiming, et al. 'Deep residual learning for image recognition.' Proceedings of the IEEE conference on computer vision and pattern recognition. 2016. [35] UyBico, Stacy J., et al. 'Lung cancer staging essentials: the new TNM staging system and potential imaging pitfalls.' Radiographics 30.5 (2010): 1163-1181. [36] Hsu, Jui-Sheng, et al. 'Pleural tags on CT scans to predict visceral pleural invasion of non–small cell lung cancer that does not abut the pleura.' Radiology 279.2 (2016): 590-596. [37] Hsu, Jui-Sheng, et al. 'Convex border of peripheral non-small cell lung cancer on CT images as a potential indicator of pleural invasion.' Medicine 96.42 (2017). [38] Iizuka, Shuhei, et al. 'A risk scoring system for predicting visceral pleural invasion in non-small lung cancer patients.' General thoracic and cardiovascular surgery 67.10 (2019): 876-879. [39] Ahn, Su Yeon, et al. 'Predictive CT features of visceral pleural invasion by T1-sized peripheral pulmonary adenocarcinomas manifesting as subsolid nodules.' American Journal of Roentgenology 209.3 (2017): 561-566. [40] Shin, Hoo-Chang, et al. 'Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning.' IEEE transactions on medical imaging 35.5 (2016): 1285-1298. [41] Choi, Hyewon, et al. 'Prediction of visceral pleural invasion in lung cancer on CT: deep learning model achieves a radiologist-level performance with adaptive sensitivity and specificity to clinical needs.' European Radiology 31.5 (2021): 2866-2876. [42] Huang, Gao, et al. 'Densely connected convolutional networks.' Proceedings of the IEEE conference on computer vision and pattern recognition. 2017. [43] Perez, Luis, and Jason Wang. 'The effectiveness of data augmentation in image classification using deep learning.' arXiv preprint arXiv:1712.04621 (2017). [44] Goodfellow, Ian, et al. 'Generative adversarial nets.' Advances in neural information processing systems 27 (2014). [45] Chuquicusma, Maria JM, et al. 'How to fool radiologists with generative adversarial networks? A visual turing test for lung cancer diagnosis.' 2018 IEEE 15th international symposium on biomedical imaging (ISBI 2018). IEEE, 2018. [46] Jarrett, Kevin, et al. 'What is the best multi-stage architecture for object recognition?.' 2009 IEEE 12th international conference on computer vision. IEEE, 2009. [47] Glorot, Xavier, Antoine Bordes, and Yoshua Bengio. 'Deep sparse rectifier neural networks.' Proceedings of the fourteenth international conference on artificial intelligence and statistics. JMLR Workshop and Conference Proceedings, 2011. [48] Jain, Anil K., Jianchang Mao, and K. Moidin Mohiuddin. 'Artificial neural networks: A tutorial.' Computer 29.3 (1996): 31-44. [49] Oktay, Ozan, et al. 'Attention u-net: Learning where to look for the pancreas.' arXiv preprint arXiv:1804.03999 (2018). [50] Vaswani, Ashish, et al. 'Attention is all you need.' Advances in neural information processing systems. 2017. [51] Hu, Jie, Li Shen, and Gang Sun. 'Squeeze-and-excitation networks.' Proceedings of the IEEE conference on computer vision and pattern recognition. 2018. [52] Xie, Saining, et al. 'Aggregated residual transformations for deep neural networks.' Proceedings of the IEEE conference on computer vision and pattern recognition. 2017. [53] Mehta, Raghav, and Jayanthi Sivaswamy. 'M-net: A convolutional neural network for deep brain structure segmentation.' 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017). IEEE, 2017. [54] Ronneberger, Olaf, Philipp Fischer, and Thomas Brox. 'U-net: Convolutional networks for biomedical image segmentation.' International Conference on Medical image computing and computer-assisted intervention. Springer, Cham, 2015. [55] 陳和豐,”基於肺部電腦斷層之肺腺癌EGFR突變預測:結合Patch-based radiomics紋理特徵圖於深度學習網路”,國立台灣大學工學院暨醫學院醫學工程學研究所,2021. [56] Sundararajan, Mukund, Ankur Taly, and Qiqi Yan. 'Axiomatic attribution for deep networks.' International Conference on Machine Learning. PMLR, 2017. [57] DeLong, Elizabeth R., David M. DeLong, and Daniel L. Clarke-Pearson. 'Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach.' Biometrics (1988): 837-845.
dc.identifier.urihttp://tdr.lib.ntu.edu.tw/jspui/handle/123456789/81948-
dc.description.abstract根據民國108年衛福部統計,癌症已連續38年位居國人十大死因之首,其中又以肺癌為兩性癌症死亡率之首,透過電腦斷層掃描(CT)之肺癌篩檢,能夠及早發現早期肺癌,進而及早進行治療,提高肺癌存活率。在早期肺癌之手術治療策略中,除了腫瘤大小外,腫瘤有無肋膜侵犯亦會影響肺癌手術術式與切除範圍。肋膜侵犯被認為是不良的預後因子,有較低的五年存活率與較高的局部復發機率,因此有肋膜侵犯的患者較不建議採取手術範圍較小之次肺葉切除術,肺葉切除術仍為此類患者的優先選擇。 臨床上判別腫瘤是否有肋膜侵犯,主要依靠CT影像上之影像特徵與臨床醫師之先驗知識,根據臨床醫師經驗不同,於判斷肋膜侵犯時會存在觀察者間之差異,且可能因疲勞增加誤判的可能性,因此,開發電腦輔助診斷系統將有助於幫助醫師判別有無肋膜侵犯,可以提供臨床醫師進行影像判斷時的第二種意見,降低觀察者間的差異,提高臨床診斷的性能與增加分類的準確性。 本研究為輔助臨床醫師進行腫瘤肋膜侵犯之判別,提出(1)基於捲積神經網路之深度學習演算法;(2)加入注意力機制於深度學習網路中,以期達到肋膜侵犯判別之理想結果,盡可能地區分出肋膜侵犯患者,擁有高靈敏度的同時,擁有良好的特異度,最大化避免非肋膜侵犯患者接受不必要之外科手術,並使深度學習網路更加專注於欲觀察之目標上。 本研究於基於捲積神經網路之深度學習演算法中,提出4 Layers Convolutional Neural Network(4L CNN)之網路架構,進行10-fold交叉驗證,可得此模型之Accuracy為0.764±0.046,Sensitivity為0.691±0.173,Specificity為0.776±0.057,於Sensitivity之表現較差。以積分梯度將深度學習模型可視化,模型之訓練結果較多關注於腫瘤邊緣、腫瘤與肺壁接觸部分等,但容易有模型關注部分於整個肺區以及肺壁部分之情況。 本研究於加入注意力機制於深度學習網路中,提出三種注意力機制: (1)Squeeze and Excitation block,(2)Dilate convolution block,(3)Lung map segmentation block,此三種注意力機制相較於單純使用4L CNN,Sensitivity皆有所進步,其中以加入Lung map segmentation block之10-fold結果及AUC表現最佳,可得模型Accuracy為0.778±0.028,Sensitivity為0.779±0.092,Specificity為0.778±0.030,AUC為0.8284。以積分梯度進行深度學習模型可視化,加入Squeeze and Excitation block之模型較容易有模型專注於整個VOI之情況,加入Dilate convolution block之模型能夠主要關注於腫瘤邊緣、肋膜標籤、肺區與肺壁之交界等,加入Lung map segmentation block之模型能夠主要關注於整體腫瘤、腫瘤邊緣、肋膜標籤、腫瘤或肺區與肺壁之交界等。 本研究使用添加Lung map segmentation block之深度學習網路架構測試外部資料,並與臨床醫師判別結果進行比較,臨床醫師之判別結果傾向Sensitivity較高,但Accuracy與Specificity較低,深度學習模型30次測試結果可得到Accuracy為0.822±0.030,Sensitivity為0.784±0.052,Specificity為0.828±0.037,雖於Sensitivity略低於臨床醫師,但Accuracy與Specificity皆高出許多。 本研究進一步將外部資料分為腫瘤有無接觸肋膜之情況,在腫瘤有接觸肋膜之情況下,深度學習模型雖於有肋膜侵犯判別之準確率略低於醫師判讀,但在無肋膜侵犯之判別高於醫師判讀,此外,醫師判讀結果落於深度學習模型ROC curve之95%信賴區間中,由實驗結果顯示,本研究提出之深度學習肋膜侵犯判別模型能與臨床醫師有相近之判別能力。zh_TW
dc.description.provenanceMade available in DSpace on 2022-11-25T03:07:14Z (GMT). No. of bitstreams: 1
U0001-0702202215284000.pdf: 4339225 bytes, checksum: 08a68611f51a2f83994a5bf441b84c77 (MD5)
Previous issue date: 2022
en
dc.description.tableofcontents論文口試委員審定書……………………………………………………………………I 致謝……………………………………………………………………………………...II 摘要....………………………………………………………………………………….III Abstract………………………………………………………………………………….V 目錄……………………………………………………………………………………VII 圖目錄………………………………………………………………………………….IX 表目錄………………………………………………………………………………….XI 第一章 緒論………………………………..……………………………..1 1.1 研究背景……………………………….….……………………..1 1.2 研究動機及目的………………………….…….………………..6 第二章 文獻回顧……………….…….…………………………………..8 2.1 影像形態學特徵……………….………………….……………..8 2.2 深度學習…………………………………….…….……………10 第三章 研究材料與方法………………………………………..………12 3.1 研究材料……………………………….…….…………………12 3.2 研究方法.......................12 3.2.1 Image preprocessing and augmentation…………..……………..13 3.2.2 4 Layers Convolutional Neural Network (4 Layers CNN)………14 3.2.3 Attention mechanism……………………………………………19 3.2.3.1 Squeeze and Excitation block……………………………………19 3.2.3.2 Dilate convolution block………………………………………...22 3.2.3.3 Lung map segmentation block…………………………………..24 3.3 Visualizing Deep Learning Model………………………………27 3.4 Performance Matrix……………………………………………..28 第四章 研究結果與討論………………………………………………..31 4.1 4 Layers Convolutional Neural Network之結果……………….31 4.2 添加Squeeze and Excitation block之結果…………………….34 4.3 添加Dilate convolution block之結果………………………….37 4.4 添加Lung map segmentation block之結果……………………40 4.5 與臨床醫師判別結果比較……………………………………..43 第五章 結論與未來展望………………………………………………..49 5.1 結論……………………………………………………………..49 5.2 未來展望………………………………………………………..50 Reference……………………………………………………………………………….52
dc.language.isozh-TW
dc.subject注意力機制zh_TW
dc.subject捲積神經網路zh_TW
dc.subject肺腫瘤肋膜侵犯分類zh_TW
dc.subject深度學習模型可視化zh_TW
dc.subject深度學習zh_TW
dc.subjectvisualization of deep learning modelen
dc.subjectClassification of visceral pleural invasion of lung tumorsen
dc.subjectdeep learningen
dc.subjectconvolutional neural networken
dc.subjectattention mechanismen
dc.title基於深度學習捲積神經網路之電腦斷層掃描肺腺癌肋膜侵犯預測模型zh_TW
dc.titlePrediction Model of Visceral Pleural Invasion in Lung Adenocarcinoma on Computed Tomography Based on Deep Learning Convolutional Neural Networken
dc.date.schoolyear110-1
dc.description.degree碩士
dc.contributor.author-orcid0000-0002-0418-7533
dc.contributor.oralexamcommittee林孟暐(Yeu-Sheng Hsieh),李佳燕(Tien-Chen Liu),(Bih-Shya Gau),(Guey-Shiun Huang)
dc.subject.keyword肺腫瘤肋膜侵犯分類,深度學習,捲積神經網路,注意力機制,深度學習模型可視化,zh_TW
dc.subject.keywordClassification of visceral pleural invasion of lung tumors,deep learning,convolutional neural network,attention mechanism,visualization of deep learning model,en
dc.relation.page56
dc.identifier.doi10.6342/NTU202200332
dc.rights.note同意授權(全球公開)
dc.date.accepted2022-02-11
dc.contributor.author-college工學院zh_TW
dc.contributor.author-dept醫學工程學研究所zh_TW
dc.date.embargo-lift2027-02-08-
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 
U0001-0702202215284000.pdf
  此日期後於網路公開 2027-02-08
4.24 MBAdobe PDF
顯示文件簡單紀錄


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。

社群連結
聯絡資訊
10617臺北市大安區羅斯福路四段1號
No.1 Sec.4, Roosevelt Rd., Taipei, Taiwan, R.O.C. 106
Tel: (02)33662353
Email: ntuetds@ntu.edu.tw
意見箱
相關連結
館藏目錄
國內圖書館整合查詢 MetaCat
臺大學術典藏 NTU Scholars
臺大圖書館數位典藏館
本站聲明
© NTU Library All Rights Reserved