請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/80070完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 陳永耀(Yung-Yaw Chen) | |
| dc.contributor.author | Yu-Shiang Lin | en |
| dc.contributor.author | 林于翔 | zh_TW |
| dc.date.accessioned | 2022-11-23T09:24:26Z | - |
| dc.date.available | 2023-07-14 | |
| dc.date.available | 2022-11-23T09:24:26Z | - |
| dc.date.copyright | 2021-07-20 | |
| dc.date.issued | 2021 | |
| dc.date.submitted | 2021-07-14 | |
| dc.identifier.citation | [1] 'Key Statistics About Liver Cancer,' Cancer.org, 2020. [Online]. Available: https://www.cancer.org/cancer/liver-cancer/about/what-is-key-statistics.html. [2] World Health Organization, 'World Cancer Report 2014,' Chapter 5.6, 2014. [3] A. Forner, J. Llovet, and J. Bruix, 'Hepatocellular carcinoma,' The Lancet, vol. 379, no. 9822, pp. 1245–1255, 2012. [4] W. Huang et al., 'Automatic HCC Detection Using Convolutional Network with Multi-Magnification Input Images,' in Proc. 2019 IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS), Taiwan, 2019, pp. 194–198. [5] Y. Liu et al., 'Detecting Cancer Metastases on Gigapixel Pathology Images,' arXiv.org, 2017. [6] Y. Li, J. Wu, and Q. Wu, 'Classification of Breast Cancer Histology Images Using Multi-Size and Discriminative Patches Based on Deep Learning,' IEEE Access, vol. 7, pp. 21400–21408, 2019. [7] Y. Wang et al., 'Classification of mice hepatic granuloma microscopic images based on a deep convolutional neural network,' Applied Soft Computing, vol. 74, pp. 40–50, 2019. [8] Y. Xu et al., 'Large scale tissue histopathology image classification, segmentation, and visualization via deep convolutional activation features,' BMC Bioinformatics, vol. 18, no. 1, 2017. [9] Y. Liu et al., 'Artificial Intelligence-Based Breast Cancer Nodal Metastasis Detection: Insights Into the Black Box for Pathologists,' Archives of Pathology Laboratory Medicine, vol. 143, no. 7, pp. 859–868, 2019. [10] B. Kiani et al., 'Impact of a deep learning assistant on the histopathologic classification of liver cancer,' npj Digital Medicine, vol. 3, no. 23, 2020. [11] M. Chen et al., 'Classification and mutation prediction based on histopathology H E images in liver cancer using deep learning,' npj Precision Oncology, vol. 4, no. 1, 2020. [12] C. Szegedy et al., 'Going Deeper with Convolutions,' arXiv.org, 2015. [13] B. E. Bejnordi et al., 'Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer,' JAMA, vol. 318, no. 22, pp. 2199-2210, 2017. [14] J. Johnson and T. Khoshgoftaar, 'Survey on deep learning with class imbalance,' Journal of Big Data, vol. 6, no. 21, 2019. [15] R. Bauder, T. Khoshgoftaar, and T. Hasanin, 'Data sampling approaches with severely imbalanced big data for medicare fraud detection,' in Proc. 2018 IEEE 30th International Conference on Tools with Artificial Intelligence (ICTAI), pp.137–42, 2018. [16] H. Lee, M. Park, and J. Kim, 'Plankton classification on imbalanced large scale database via convolutional neural networks with transfer learning,' in Proc. 2016 IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, USA, pp. 3713–7, 2016. [17] M. Koziarski, B. Kwolek, and B. Cyganek, 'Convolutional neural network-based classification of histopathological images affected by data imbalance,' in Video Analytics. Face and Facial Expression Recognition, Springer, pp. 1–11, 2018. [18] U. Michelucci, 'How to Split Your Dataset,' in Applied Deep Learning, C. S. John, Ed., New York, NY, USA: Apress, 2018, pp. 230–234. [19] R. Figueroa, Q. Zeng-Treitler, S. Kandula, and L. Ngo, 'Predicting sample size required for classification performance,' BMC Medical Informatics and Decision Making, vol. 12, no. 1, 2012. [20] S. Mukherjee et al., 'Estimating Dataset Size Requirements for Classifying DNA Microarray Data,' Journal of Computational Biology, vol. 10, no. 2, pp. 119–142, 2003. [21] J. Cho, K. Lee, E. Shin, G. Choy, and S. Do, 'How much data is needed to train a medical image deep learning system to achieve necessary high accuracy?' arXiv.org, 2015. [22] A. Damien, 'TFLearn | TensorFlow Deep Learning Library,' Tflearn.org, 2020. [Online]. Available: http://tflearn.org/. [23] E. Wilson, 'Probable Inference, the Law of Succession, and Statistical Inference,' Journal of the American Statistical Association, vol. 22, no. 158, pp. 209–212, 1927. [24] Echle, A., Rindtorff, N.T., Brinker, T.J. et al. 'Deep learning in cancer pathology: a new generation of clinical biomarkers,' Br J Cancer, vol. 124, pp. 686–696, 2021. [25] C. Chen, Y. Huang, P. Fang, C. Liang, and R. Chang, 'A computer-aided diagnosis system for differentiation and delineation of malignant regions on whole-slide prostate histopathology image using spatial statistics and multidimensional DenseNet,' Med Phys, vol. 47(3), pp. 1021–1033, 2020. [26] D. Wang, A. Khosla, R. Gargeya, H. Irshad, and A. H. Beck, 'Deep learning for identifying metastatic breast cancer,' arXiv preprint arXiv:1606.05718, 2016. [27] CAMELYON16 Challenge. http://camelyon16.grand-challenge.org/ [28] MICCAI 2014 Brain Tumor Digital Pathology Challenge. https://wiki.cancerimagingarchive.net/display/Public/MICCAI+2014+Grand+Challenges [29] A. Krizhevsky, I. Sutskever, and GE. Hinton, 'Imagenet classification with deep convolutional neural networks,' Advances in neural information processing systems, vol. 25, pp. 1097–1105, 2012. [30] The Cancer Genome Atlas (TCGA). https://www.cancer.gov/about-nci/organization/ccg/research/structural-genomics/tcga [31] S. Ramaswamy, R.T. Osteen, and L.N. Shulman, 'Metastatic cancer from an unknown primary site,' Clinical Oncology, pp. 711–719, 2001. [32] UCI Machine Learning Repository. http://www.ics.uci.edu/~mlearn/MLRepository.html [33] F.A. Spanhol, L.S. Oliveira, C. Petitjean, L. Heutte, 'A dataset for breast cancer histopathological image classification,' IEEE Transactions on Biomedical Engineering, vol. 63, no. 7, pp. 1455-1462, 2016. [34] MM. Mukaka., 'Statistics corner: A guide to appropriate use of correlation coefficient in medical research,' Malawi Med J., vol. 24, no. 3, pp. 69–71, 2012. [35] Y. Lin, P. Huang and Y. Chen, 'Deep Learning-Based Hepatocellular Carcinoma Histopathology Image Classification: Accuracy Versus Training Dataset Size,' IEEE Access, vol. 9, pp. 33144-33157, 2021. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/80070 | - |
| dc.description.abstract | "在全球，肝癌每年導致700,000多例死亡，是癌症死亡的第二大主要原因。肝細胞癌（Hepatocellular carcinoma, HCC）是成人中最常見的肝癌類型，占肝硬化患者的大多數死亡原因。早期的肝癌患者如果可以透過手術干預治療，預後通常會較為良好。因此，早期的病理影像學診斷是對抗肝癌的必要步驟。然而，常規的人工病理學診斷，病理醫師需要耗費大量的時間和精力，針對病理影像中的癌細胞正確位置進行詳細的檢查，並且，不同經驗的醫師的檢查結果可能也會有所差異，這對於診斷的準確率以及效率帶來了極大的挑戰。近年來，基於深度學習的病理影像分類器，在其他不同類型的病理影像的分類研究中，已被證明可以有效的輔助病理醫師進行更快速且正確的診斷。 根據過往研究指出，深度學習的分類準確率，與已標記的訓練資料數量呈現正相關。然而，通常難以確定需要標記多少數量的病理影像做為訓練資料，才能在臨床診斷中取得良好的表現。值得注意的是，一張病理影像的尺寸通常高達數十億像素以上，此特殊性質使得病理影像的人工標記成本相對於其他醫學影像昂貴許多。因此，如果在進行標記工作之前，就能夠依照診斷需求，提前估計所需要的訓練資料數量，應可更有效的優化標記成本、降低人力負擔。然而，目前對於病理影像所需要的訓練資料數量估計，仍缺乏可參考的相關研究。因此，本研究除了運用深度學習方法，對於肝細胞癌病理影像進行二分類，亦深入探討了肝細胞癌病理影像的分類準確率，與用於訓練的已標記資料集大小之間的關係，本研究的主要貢獻如下: 第一、本研究應用了GoogLeNet (Inception-V1)深度學習模型，對於肝細胞癌病理影像進行二分類。本研究運用了25張個案影像所訓練之模型，對於未經過訓練之4張個案影像進行分類測試，其分類準確率可達91.37%（±2.49%），靈敏度可達92.16%（±4.93%），特異性可達90.57%（±2.54%）。除此之外，本研究透過了單一影像所訓練之模型，確定了肝細胞癌病理影像的多樣性，將會極大程度的影響模型對於未經過訓練之個案影像的分類準確率。 第二、本研究深入探討了不同大小的訓練影像資料集，與其對應之未經過訓練之個案影像分類準確率的關係。基於此關係，本研究進一步運用了一種基於逆冪函數的估計模型 (Inverse power law function-based estimation model)，預估到達臨床診斷所需的分類準確率時，所需要標記的肝細胞癌病理影像的最低數量，此估計數量可作為未來肝細胞癌病理影像訓練資料收集、以及標記的重要參考依據。 第三、本研究深入探討了不同數量的肝細胞癌病理影像訓練子切片(Patch)，與其對應之訓練影像分類準確率的關係。於此基礎之上，本研究進一步提出了一種基於低信心率的估計方法 (Low Confidence Rate-based estimation method)，本方法可依據所需要的分類準確率，預估每一張不同的肝細胞癌病理影像，所需要標記的訓練子切片最低數量，此估計值可作為病理醫師標記肝細胞癌病理影像子切片時的最低數量參考依據。" | zh_TW |
| dc.description.provenance | Made available in DSpace on 2022-11-23T09:24:26Z (GMT). No. of bitstreams: 1 U0001-1207202121304000.pdf: 6100523 bytes, checksum: ec84f119fb52892aa633a89a80c0f182 (MD5) Previous issue date: 2021 | en |
| dc.description.tableofcontents | "口試委員會審定書 i 誌謝 ii 中文摘要 iii Abstract v Contents vii List of Figures x List of Tables xx Chapter 1 Introduction 1 1.1 Motivation 1 1.2 Problem Definition 2 1.3 Previous Approach 3 1.4 Proposed Approach 4 1.5 Thesis Overview 5 Chapter 2 Review of Related Work 6 2.1 Deep Learning-Based Histopathology Image Classification 9 2.1.1 Breast Tumor Classification 9 2.1.2 Brain Tumor Classification 13 2.1.3 Prostate Tumor Classification 15 2.2 Deep Learning-Based HCC Histopathology Image Classification 19 2.2.1 HCC vs. CC Classification 20 2.2.2 HCC vs. Normal Classification 23 2.3 Annotated Training Dataset Size Estimation 26 2.3.1 Inverse Power Law Function-Based Estimation 27 2.3.2 Training Dataset Size Estimation for Deep Learning 34 2.4 Summary and Comparison 37 Chapter 3 HCC Histopathology Image Classification 44 3.1 Proposed Method for HCC Histopathology Image Classification 44 3.2 Class-Imbalance Problem of Image Classification 49 3.2.1 Class-Imbalance Problem of Large-Scale Dataset 49 3.2.2 Class-Imbalance Problem of Histopathology Image 50 3.3 Random Under-Sampling Method 51 Chapter 4 Training Dataset Size Estimation for Desired Accuracy 53 4.1 Training with Different Number of Images 53 4.2 Inverse Power Law Function-Based Estimation for Training Images 55 4.3 Training with Different Number of Patches 56 4.4 LCR-Based Estimation for Training Patches 58 Chapter 5 Dataset of HCC Histopathology Image 61 5.1 Patches Preparation 63 5.2 Number of Patches 65 Chapter 6 Experiment and Discussions 67 6.1 Evaluation of 25-Image Training Model 68 6.2 Testing Accuracies for Different Training Dataset Sizes 72 6.2.1 Single-Image Training 72 6.2.2 2-Image Training 76 6.2.3 13-Image Training 78 6.2.4 25-Image Training 80 6.3 Variation of Overall-TATI and Overall-TANI 83 6.4 Required Number of Training Image Estimation 85 6.5 Required Number of Training Patches Estimation 88 6.5.1 Number of Training Patches vs. MAE and TATI 88 6.5.2 Correlation between LCR, MAE, and TATI 96 6.5.3 Proposed LCR-based Estimation Method 106 6.6 Discussions and Summary 112 Chapter 7 Conclusions and Future Work 114 References 116" | |
| dc.language.iso | en | |
| dc.subject | 低信心率估計方法 | zh_TW |
| dc.subject | 卷積神經網路 | zh_TW |
| dc.subject | 深度學習 | zh_TW |
| dc.subject | 肝細胞癌 | zh_TW |
| dc.subject | 組織病理學圖像分類 | zh_TW |
| dc.subject | 逆冪函數擬合曲線 | zh_TW |
| dc.subject | deep learning | en |
| dc.subject | Low Confidence Rate-based estimation method | en |
| dc.subject | inverse power law function-based fitting curve | en |
| dc.subject | histopathology image classification | en |
| dc.subject | Convolutional neural network | en |
| dc.subject | hepatocellular carcinoma | en |
| dc.title | 基於深度學習之HCC病理影像分類:準確率與訓練資料集大小之關係 | zh_TW |
| dc.title | Deep Learning-Based Hepatocellular Carcinoma Histopathology Image Classification: Accuracy versus Training Dataset Size | en |
| dc.date.schoolyear | 109-2 | |
| dc.description.degree | 博士 | |
| dc.contributor.oralexamcommittee | 顏家鈺(Hsin-Tsai Liu),傅立成(Chih-Yang Tseng),張智星,黃佩欣 | |
| dc.subject.keyword | 卷積神經網路,深度學習,肝細胞癌,組織病理學圖像分類,逆冪函數擬合曲線,低信心率估計方法, | zh_TW |
| dc.subject.keyword | Convolutional neural network,deep learning,hepatocellular carcinoma,histopathology image classification,inverse power law function-based fitting curve,Low Confidence Rate-based estimation method, | en |
| dc.relation.page | 118 | |
| dc.identifier.doi | 10.6342/NTU202101419 | |
| dc.rights.note | 同意授權(全球公開) | |
| dc.date.accepted | 2021-07-15 | |
| dc.contributor.author-college | 電機資訊學院 | zh_TW |
| dc.contributor.author-dept | 電機工程學研究所 | zh_TW |
| dc.date.embargo-lift | 2023-07-14 | - |
| 顯示於系所單位： | 電機工程學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| U0001-1207202121304000.pdf | 5.96 MB | Adobe PDF | 檢視/開啟 |
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。