Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/85656Full metadata record
| ???org.dspace.app.webui.jsptag.ItemTag.dcfield??? | Value | Language |
|---|---|---|
| dc.contributor.advisor | 陳中明(Chung-Ming Chen) | |
| dc.contributor.author | CHING-CHIA CHUANG | en |
| dc.contributor.author | 莊靜佳 | zh_TW |
| dc.date.accessioned | 2023-03-19T23:20:44Z | - |
| dc.date.copyright | 2022-07-05 | |
| dc.date.issued | 2022 | |
| dc.date.submitted | 2022-06-24 | |
| dc.identifier.citation | [1] 衛生福利部109年國人死因統計結果 [2] Travis, W. D., Brambilla, E., Noguchi, M., Nicholson, A. G., Geisinger, K. R., Yatabe, Y., Beer, D. G., Powell, C. A., Riely, G. J., & Van Schil, P. E. (2011). International association for the study of lung cancer/american thoracic society/european respiratory society international multidisciplinary classification of lung adenocarcinoma. Journal of Thoracic Oncology, 6(2), 244-285. [3] Liu, S., Wang, R., Zhang, Y., Li, Y., Cheng, C., Pan, Y., Xiang, J., Zhang, Y., Chen, H., & Sun, Y. (2016). Precise diagnosis of intraoperative frozen section is an effective method to guide resection strategy for peripheral small-sized lung adenocarcinoma. Journal of clinical oncology, 34(4), 307-313. [4] Nesbitt, J. C., Putnam Jr, J. B., Walsh, G. L., Roth, J. A., & Mountain, C. F. (1995). Survival in early-stage non-small cell lung cancer. The Annals of thoracic surgery, 60(2), 466-472. [5] Jemal, A., Siegel, R., Xu, J., & Ward, E. (2010). Cancer statistics, 2010. CA: a cancer journal for clinicians, 60(5), 277-300. [6] Detterbeck, F. C. (2018). The eighth edition TNM stage classification for lung cancer: What does it mean on main street? The Journal of thoracic and cardiovascular surgery, 155(1), 356-359. [7] Travis, W. D., Asamura, H., Bankier, A. A., Beasley, M. B., Detterbeck, F., Flieder, D. B., Goo, J. M., MacMahon, H., Naidich, D., & Nicholson, A. G. (2016). The IASLC lung cancer staging project: proposals for coding T categories for subsolid nodules and assessment of tumor size in part-solid tumors in the forthcoming eighth edition of the TNM classification of lung cancer. Journal of Thoracic Oncology, 11(8), 1204-1223. [8] Fan, L., Fang, M., Li, Z., Tu, W., Wang, S., Chen, W., Tian, J., Dong, D., & Liu, S. (2019). Radiomics signature: a biomarker for the preoperative discrimination of lung invasive adenocarcinoma manifesting as a ground-glass nodule. European radiology, 29(2), 889-897. [9] Cho, H.-h., Lee, G., Lee, H. Y., & Park, H. (2020). Marginal radiomics features as imaging biomarkers for pathological invasion in lung adenocarcinoma. European radiology, 30(5), 2984-2994. [10] Wu, L., Gao, C., Xiang, P., Zheng, S., Pang, P., & Xu, M. (2020). CT-imaging based analysis of invasive lung adenocarcinoma presenting as ground glass nodules using peri-and intra-nodular radiomic features. Frontiers in oncology, 10, 838. [11] Wu, G., Woodruff, H. C., Shen, J., Refaee, T., Sanduleanu, S., Ibrahim, A., Leijenaar, R. T., Wang, R., Xiong, J., & Bian, J. (2020). Diagnosis of invasive lung adenocarcinoma based on chest CT radiomic features of part-solid pulmonary nodules: a multicenter study. Radiology, 297(2), 451-458. [12] Gong, J., Liu, J., Hao, W., Nie, S., Zheng, B., Wang, S., & Peng, W. (2020). A deep residual learning network for predicting lung adenocarcinoma manifesting as ground-glass nodule on CT images. European radiology, 30(4), 1847-1855. [13] Xia, X., Gong, J., Hao, W., Yang, T., Lin, Y., Wang, S., & Peng, W. (2020). Comparison and fusion of deep learning and radiomics features of ground-glass nodules to predict the invasiveness risk of stage-I lung adenocarcinomas in CT scan. Frontiers in oncology, 418. [14] Suzuki, K., Koike, T., Asakawa, T., Kusumoto, M., Asamura, H., Nagai, K., Tada, H., Mitsudomi, T., Tsuboi, M., & Shibata, T. (2011). A prospective radiological study of thin-section computed tomography to predict pathological noninvasiveness in peripheral clinical IA lung cancer (Japan Clinical Oncology Group 0201). Journal of Thoracic Oncology, 6(4), 751-756. [15] Suzuki, K., Watanabe, S.-i., Wakabayashi, M., Saji, H., Aokage, K., Moriya, Y., Yoshino, I., Tsuboi, M., Nakamura, S., & Nakamura, K. (2022). A single-arm study of sublobar resection for ground-glass opacity dominant peripheral lung cancer. The Journal of thoracic and cardiovascular surgery, 163(1), 289-301. e282. [16] Freund, Y., & Schapire, R. E. (1997). A decision-theoretic generalization of on-line learning and an application to boosting. Journal of computer and system sciences, 55(1), 119-139. [17] Duda, R. O., Hart, P. E., & Stork, D. G. (2000). Pattern Classification. John Wiley & Sons. Inc., New York, 5. [18] Ye, R., Liu, F., & Zhang, L. (2019). 3D depthwise convolution: reducing model parameters in 3D vision tasks. Canadian Conference on Artificial Intelligence, [19] Lin, M., Chen, Q., & Yan, S. (2013). Network in network. arXiv preprint arXiv:1312.4400. [20] Hertz, J., Krogh, A., & Palmer, R. G. (2018). Introduction to the theory of neural computation. CRC Press. [21] Glorot, X., Bordes, A., & Bengio, Y. (2011). Deep sparse rectifier neural networks. Proceedings of the fourteenth international conference on artificial intelligence and statistics, [22] Hu, J., Shen, L., & Sun, G. (2018). Squeeze-and-excitation networks. Proceedings of the IEEE conference on computer vision and pattern recognition, [23] Taherkhani, A., Cosma, G., & McGinnity, T. M. (2020). AdaBoost-CNN: An adaptive boosting algorithm for convolutional neural networks to classify multi-class imbalanced datasets using transfer learning. Neurocomputing, 404, 351-366. [24] Hastie, T., Rosset, S., Zhu, J., & Zou, H. (2009). Multi-class adaboost. Statistics and its Interface, 2(3), 349-360. [25] Lim, H.-j., Ahn, S., Lee, K. S., Han, J., Shim, Y. M., Woo, S., Kim, J.-H., Yie, M., Lee, H. Y., & Chin, A. Y. (2013). Persistent pure ground-glass opacity lung nodules≥ 10 mm in diameter at CT scan: histopathologic comparisons and prognostic implications. Chest, 144(4), 1291-1299. [26] Grove, O., Berglund, A. E., Schabath, M. B., Aerts, H. J., Dekker, A., Wang, H., Velazquez, E. R., Lambin, P., Gu, Y., & Balagurunathan, Y. (2015). Quantitative computed tomographic descriptors associate tumor shape complexity and intratumor heterogeneity with prognosis in lung adenocarcinoma. PloS one, 10(3), e0118261. [27] Son, J. Y., Lee, H. Y., Lee, K. S., Kim, J.-H., Han, J., Jeong, J. Y., Kwon, O. J., & Shim, Y. M. (2014). Quantitative CT analysis of pulmonary ground-glass opacity nodules for the distinction of invasive adenocarcinoma from pre-invasive or minimally invasive adenocarcinoma. PloS one, 9(8), e104066. [28] Zhang, Y., Heuvelmans, M., Zhang, H., Oudkerk, M., Zhang, G., & Xie, X. (2018). Changes in quantitative CT image features of ground-glass nodules in differentiating invasive pulmonary adenocarcinoma from benign and in situ lesions: histopathological comparisons. Clinical radiology, 73(5), 504. e509-504. e516. [29] Chae, H.-D., Park, C. M., Park, S. J., Lee, S. M., Kim, K. G., & Goo, J. M. (2014). Computerized texture analysis of persistent part-solid ground-glass nodules: differentiation of preinvasive lesions from invasive pulmonary adenocarcinomas. Radiology, 273(1), 285-293. [30] Mitchell, M. (1998). An introduction to genetic algorithms. MIT press. [31] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25. [32] Sagi, O., & Rokach, L. (2018). Ensemble learning: A survey. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 8(4), e1249. [33] Chen, L.-C., Papandreou, G., Kokkinos, I., Murphy, K., & Yuille, A. L. (2017). Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs. IEEE transactions on pattern analysis and machine intelligence, 40(4), 834-848. [34] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition, [35] Iandola, F. N., Han, S., Moskewicz, M. W., Ashraf, K., Dally, W. J., & Keutzer, K. (2016). SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5 MB model size. arXiv preprint arXiv:1602.07360. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/85656 | - |
| dc.description.abstract | 根據衛生福利部於109年國人死因統計結果[1]，癌症的死亡率在十大死因中排序第一，而又以氣管、支氣管、肺癌位於癌症中的第一且已蟬聯多年，是一項不可輕忽的疾病，若能及早給予患者相應的治療的話，可以相對提高存活率。較早期的肺部腫瘤在電腦斷層掃描上通常會呈現毛玻璃狀，毛玻璃狀定義即為腫瘤呈現霧狀並且不會遮擋住穿越之血管或是蓋住氣管等組織。另外，於2011年有三個協會：國際肺癌協會（IASLC）、美國胸科協會（ATS）及歐洲呼吸協會（ERS），共同從病理角度針對毛玻璃狀肺腺癌做分類，以他們對人體所造成的威脅程度可以分成三大類，分別為較不具威脅性的pre-invasive lesions，包含非典型腺瘤樣增生（AAH）與原位肺腺癌（AIS），以及另外兩類：微浸潤腺癌（MIA）和浸潤腺癌（IA）[2]，而不同分類的肺腺癌所需的開刀方式與存活率也不盡相同，其中AIS/MIA有些醫生的準則會選擇先以追蹤為主，或使用次肺葉切除術（楔狀切除術或是肺節切除術），在術後五年的存活率是接近100%，反觀浸潤腺癌（IA）的存活率則有所降低（依照不同的亞型而有不同的存活率）[3]。所以若能從電腦斷層掃描上將早期毛玻璃狀腫瘤中的原位肺腺癌與微浸潤線癌從浸潤腺癌中分出來可以給予醫生應進行手術還是應先觀察的一個參考，是相當重要的事。 於本論文，將採用實質比小於0.25且小於三公分的毛玻璃狀原位肺腺癌（AIS）、微浸潤線癌（MIA）與浸潤腺癌（IA）當作研究材料，經醫師判斷後其須符合腫瘤實質區域之最大直徑與腫瘤本體之最大徑比值小於0.25，並將AIS與MIA歸類為不具侵犯性的肺腺癌，而IA屬於具侵犯性的肺腺癌。研究目的為利用放射體學與深度學習的方法將具侵犯性的肺腺癌與不具侵犯性的肺腺癌給區分出來。 | zh_TW |
| dc.description.abstract | According to the report from Ministry of the health and welfare in 2020[1], cancer ranked first in the cause of death statistics. Among all cancer cause of death, cancers of trachea, bronchus and lung were placed first for a long time. Apparently, it’s important to have an early detection in order to lead to cure and enhance the survival rate since lung cancer is a force to be reckoned with. Early-stage lung adenocarcinoma nodules often manifest as ground-glass opacity (GGO) which is defined as lesions showing hazy, increased attenuation that does not obscure underlying bronchial structures or pulmonary vessels. In 2011, the International Association for the Study of Lung Cancer (IASLC), the American Thoracic Society(ATS), and the European Respiratory Society (ERS) classified lung adenocarcinomas manifest as GGO into three groups of type:（1）pre-invasive lesions, including atypical adenomatous hyperplasias (AAH) and adenocarcinoma on situ (AIS), （2）minimally invasive adenocarcinoma (MIA), and（3）invasive adenocarcinoma (IA)[2]. The lung adenocarcinomas from each groups are suggested for different therapeutic strategy. AIS and MIA can be tracked at first or treated with sublobar resection (wedge or segmental resection) with a 100% or nearly 100% of 5-year survival-rate[3]. On the other hand, the invasive adenocarcinoma causes a reduction in survival-rate (Which depends on the subtype of the adenocarcinoma). As stated above, classifying AIS and MIA from IA which manifest as GGO on computed tomography is crucial that either gives the doctors an option to track first or to perform the operation. In this study, the inclusion criteria are the maximum diameter is less than 3 cm and the solid ratio of the ground-glass nodules must be less than 0.25 which judged by the doctor, indicates the ratio of max diameter of the solid lesion to max diameter of the whole lesion need to be less than 0.25. Furthermore, AIS and MIA are classified as non-invasive adenocarcinoma while IAs are invasive adenocarcinoma. The purpose of the study is to use radiomics and deep learning to build up classification model so as to make a precise precision to the classification problem. | en |
| dc.description.provenance | Made available in DSpace on 2023-03-19T23:20:44Z (GMT). No. of bitstreams: 1 U0001-2106202220090800.pdf: 5698661 bytes, checksum: 8e4a445d20b1765747fe1187877be6d8 (MD5) Previous issue date: 2022 | en |
| dc.description.tableofcontents | 摘要 II Abstract III 圖目錄 VI 表目錄 VIII 第一章 緒論 1 1.1 研究背景 1 1.2 肺癌分類與分期介紹 3 1.3 AIS, MIA, IA 定義 5 1.4 研究動機與目的 8 第二章 文獻探討 13 2.1 毛玻璃狀腫瘤之侵犯性分類—放射體學模型 13 2.2 毛玻璃狀腫瘤之侵犯性分類—深度學習 17 2.3 毛玻璃狀腫瘤之侵犯性分類—放射體學結合深度學習 18 第三章 研究方法與材料 21 3.1 研究材料 21 3.2 研究方法 22 3.2.1. Adaboost演算法 23 3.2.2. 影像處理 24 3.2.3. 模型架構 28 3.2.4. 深度模型更新資料與權重 37 3.3 效能評估 41 3.3.1 效能指標 41 3.3.2 比較文獻介紹之方法 44 3.3.3 本研究方法之比較 65 第四章 研究結果與討論 77 4.1 針對分不好的資料進行著重學習的三種方式 77 4.2 kernel number數量的影響 78 4.3 Attention layer的重要性 79 4.4 影像輸入 80 4.5 fusion 81 4.6 Solid ratio的影響 82 第五章 結論與未來展望 85 5.1 結論與未來展望 85 5.2 研究限制 86 參考資料 87 | |
| dc.language.iso | zh-TW | |
| dc.subject | 放射體學與深度學習 | zh_TW |
| dc.subject | 肺部電腦斷層掃描影像 | zh_TW |
| dc.subject | 毛玻璃樣腫瘤 | zh_TW |
| dc.subject | 低實質比 | zh_TW |
| dc.subject | 毛玻璃樣腫瘤是否具侵犯性的分類 | zh_TW |
| dc.subject | low solid ratio | en |
| dc.subject | ground glass nodule | en |
| dc.subject | Lung computed tomography scan | en |
| dc.subject | classification of the invasiveness of the ground glass nodules | en |
| dc.subject | radiomics and deep learning | en |
| dc.title | 肺部電腦斷層掃描之毛玻璃狀肺結節侵犯性區別診斷:低實質比、深度學習與放射體學方法 | zh_TW |
| dc.title | Classification of the Invasiveness of Ground Glass Nodule on CT images: low solid ratio, deep learning and radiomics method | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 110-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 李佳燕(Chia-Yen Lee),陳正剛(Argon Chen),林孟暐(Mong-Wei Lin) | |
| dc.subject.keyword | 肺部電腦斷層掃描影像,毛玻璃樣腫瘤,低實質比,毛玻璃樣腫瘤是否具侵犯性的分類,放射體學與深度學習, | zh_TW |
| dc.subject.keyword | Lung computed tomography scan,ground glass nodule,low solid ratio,classification of the invasiveness of the ground glass nodules,radiomics and deep learning, | en |
| dc.relation.page | 91 | |
| dc.identifier.doi | 10.6342/NTU202201042 | |
| dc.rights.note | 同意授權(全球公開) | |
| dc.date.accepted | 2022-06-27 | |
| 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-lift | 2022-07-05 | - |
| Appears in Collections: | 醫學工程學研究所 | |
Files in This Item:
| File | Size | Format | |
|---|---|---|---|
| U0001-2106202220090800.pdf | 5.57 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.