請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/89979完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 曾雪峰 | zh_TW |
| dc.contributor.advisor | Snow H. Tseng | en |
| dc.contributor.author | 張庭宇 | zh_TW |
| dc.contributor.author | Ting-Yu Chang | en |
| dc.date.accessioned | 2023-09-22T16:55:15Z | - |
| dc.date.available | 2023-11-09 | - |
| dc.date.copyright | 2023-09-22 | - |
| dc.date.issued | 2023 | - |
| dc.date.submitted | 2023-08-13 | - |
| dc.identifier.citation | 馮昱豪 and 陳浩民, <以電腦輔助診斷技術為核心之智慧醫療專利技術趨勢分析>. 智慧財產權月刊 VOL.258, 2020.
Doi, K., Computer-aided diagnosis in medical imaging: historical review, current status and future potential. Computerized medical imaging and graphics, 2007. 31(4-5): p. 198-211. Huang, D., et al., Optical coherence tomography. science, 1991. 254(5035): p. 1178-1181. Ogawa, S., et al., Brain magnetic resonance imaging with contrast dependent on blood oxygenation. proceedings of the National Academy of Sciences, 1990. 87(24): p. 9868-9872. Leitgeb, R., C. Hitzenberger, and A.F. Fercher, Performance of fourier domain vs. time domain optical coherence tomography. Optics express, 2003. 11(8): p. 889-894. Tsai, C.-C., et al., Full-depth epidermis tomography using a Mirau-based full-field optical coherence tomography. Biomedical Optics Express, 2014. 陳昱彤, 全域式光學同調斷層掃描術用於動物眼睛模型之特性分析. 2018. Schmidhuber, J., Deep learning in neural networks: An overview. Neural networks, 2015. 61: p. 85-117. Radford, A., L. Metz, and S. Chintala, Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434, 2015. Miyato, T., et al., Virtual adversarial training: a regularization method for supervised and semi-supervised learning. IEEE transactions on pattern analysis and machine intelligence, 2018. 41(8): p. 1979-1993. Littman, M. and A. Moore, Reinforcement Learning: A Survey, Journal of Artificial Intelligence Research 4. 1996, syf. Ronneberger, O., P. Fischer, and T. Brox. U-net: Convolutional networks for biomedical image segmentation. in International Conference on Medical image computing and computer-assisted intervention. 2015. Springer. | - |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/89979 | - |
| dc.description.abstract | 本文中我們使用全域式光學同調斷層掃描得到的三圍老鼠皮膚病變影像作為資料集，並利用機器學習的原理，建構出能用於皮膚病變診斷的分類模型，評估各個模型後對模型架構進行修改，選擇出診斷準確率最好的CNN模型，並且對模型的感受野參數進行更深入的研究，探討感受野的變化對模型準確率的影響，挑選出最適合的參數數值與模型架構，最終得到81.43%的準確率，可知本模型在皮膚癌的臨床診斷上具備一定的可行性。 | zh_TW |
| dc.description.abstract | In this study, we utilized three-dimensional skin lesion images obtained through full-field optical coherence tomography (FF-OCT) as our dataset. Leveraging the principles of machine learning, we constructed a classification model for diagnosing skin lesions. After evaluating various models, we made modifications to the model structure. The CNN model with the highest accuracy was selected. Furthermore, we conducted an in-depth investigation into the parameters of the model. We explored the impact of changes in receptive field on model accuracy, identifying the most suitable parameter values and model structure. As a result, we achieved an accuracy of 81.43%. This suggests that our model demonstrates a certain level of feasibility for clinical diagnosis of skin lesions and skin cancer. | en |
| dc.description.provenance | Submitted by admin ntu (admin@lib.ntu.edu.tw) on 2023-09-22T16:55:15Z No. of bitstreams: 0 | en |
| dc.description.provenance | Made available in DSpace on 2023-09-22T16:55:15Z (GMT). No. of bitstreams: 0 | en |
| dc.description.tableofcontents | 口試委員會審定書 #
誌謝 i 中文摘要 ii ABSTRACT iii 目錄 iv 圖目錄 vii 表目錄 ix Chapter 1 緒論 1 1.1 研究動機 1 1.2 研究目的 2 1.3 論文架構 2 Chapter 2 研究背景 3 2.1 智慧醫療技術 3 2.2 光學同調斷層掃描 3 2.3 人工智慧與深度學習 6 2.4 卷積神經網路 9 2.4.1 卷積層 10 2.4.2 池化層 12 2.4.3 平坦化 13 2.4.4 完全連接層 14 Chapter 3 研究方法 17 3.1 老鼠皮膚之光學同調掃描影像 17 3.2 預處理 20 3.2.1 影像預處理 22 3.2.2 標籤化 24 3.3 機器學習模型設計 25 3.3.1 LeNet 25 3.3.2 ResNet 26 3.3.3 AlexNet 28 3.3.4 VGGNet 29 3.4 深度學習模型 31 3.5 超參數配置 31 3.5.1 權重初始化 31 3.5.2 學習速率 32 3.5.3 迭代週期與批次大小 33 3.5.4 優化器 33 3.6 評估與量化指標 37 3.6.1 損失函數 37 3.6.2 過擬合與欠擬合 38 3.6.3 評估指標 40 3.6.4 交叉驗證 43 Chapter 4 研究結果 46 4.1 診斷結果驗證 46 4.1.1 Holdout驗證 46 4.1.2 K-fold交叉驗證 51 4.2 可視化分析 51 4.2.1 Grad-CAM 51 4.2.2 感受野 (Receptive Field)調整對學習成果的影響 54 Chapter 5 結論 59 5.1 研究結論 59 5.2 未來展望 59 參考資料 61 | - |
| dc.language.iso | zh_TW | - |
| dc.title | 人工智慧在醫學影像上之應用與模型參數分析 | zh_TW |
| dc.title | Applications of Artificial Intelligence in Medical Imaging and Analysis of Parameters | en |
| dc.type | Thesis | - |
| dc.date.schoolyear | 111-2 | - |
| dc.description.degree | 碩士 | - |
| dc.contributor.oralexamcommittee | 蕭惠心;黃定洧 | zh_TW |
| dc.contributor.oralexamcommittee | Hui-Hsin Hsiao;Ding-Wei Huang | en |
| dc.subject.keyword | 全域式光學同調斷層掃描,機器學習,人工智慧,圖像辨識,卷積神經網路,光學影像, | zh_TW |
| dc.subject.keyword | FF-OCT,Machine learning,Optical images,Artificial intelligence,CNN, | en |
| dc.relation.page | 62 | - |
| dc.identifier.doi | 10.6342/NTU202304105 | - |
| dc.rights.note | 未授權 | - |
| dc.date.accepted | 2023-08-14 | - |
| dc.contributor.author-college | 電機資訊學院 | - |
| dc.contributor.author-dept | 光電工程學研究所 | - |
| 顯示於系所單位： | 光電工程學研究所 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-111-2.pdf 目前未授權公開取用 | 3.62 MB | Adobe PDF |
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。