臺灣博碩士論文加值系統

光學同調斷層掃描(Optical Coherence Tomography,OCT) 是一種獲取與處理光學信號的成像技術，利用低同調光進行掃描，從樣本內部拍攝微米級解析度的二維或三維圖像，它已應用於許多領域，如:醫學中的人體組織成像和工業中的缺陷檢測等等，透過雷射光進入樣本內部後反射回來的光訊號進行解調來獲得樣本內部構造的資訊，而且它具有精準、快速、穩定性、和非侵入式的優點。
然而掃描的樣本資訊可能會受到不同假影雜訊的影響，使掃描的準確性降低，所以本論文我們將OCT的技術和深度學習的Pix2Pix生成對抗網路進行結合，透過使用Pix2PixGAN 進行訓練和驗證，來改善掃描圖像的品質。本論文包含兩個實驗，第一個實驗為將圖像中的鏡像雜訊消除，第二個實驗則是將圖像中受到振動干擾而產生的移動假影消除，將雜訊消除後再生成清晰的正常掃描圖像，最後會探討訓練數據量的多寡和訓練次數對生成圖像品質的影響。

Optical Coherence Tomography (OCT) is an imaging technique that acquires and processes optical signals, using low-coherence light for scanning to capture two or three dimensional images with micrometer-level resolution from samples. It has been applied in various fields such as medical imaging for human tissue and industrial defect detection.Using laser light directed into the sample and analyzing the reflected signal, OCT obtains information about the internal structure of the sample. It has advantages of precision, speed,stability, and non-invasiveness.
However, the scanned sample information may be affected by various artifacts and noise, leading to a decrease in accuracy. Therefore, in this paper, we combine OCT technology with the deep learning Pix2Pix Generative Adversarial Network (GAN) to improve the quality of images.The study includes two experiments. the first experiment aims to eliminate mirror noise in the images,the second experiment focuses on eliminating motion artifacts caused by vibrational interference. After noise elimination, the images are reconstructed into normal images. Finally, the paper explores the impact of the amount of training data and the number of training iterations on the quality of images.

摘要.....................i
Abstract.................ii
誌謝......................iii
目錄......................iv
表目錄....................vi
圖目錄....................vii
第一章 緒論................1
1.1前言...................1
1.2研究動機和目的..........2
1.3論文架構...............3
1.4文獻探討...............3
第二章 光學同調斷層掃描技術背景與原理介紹...............7
2.1簡介.............................................7
2.2 OCT干涉原理......................................8
2.3時域光學同調斷層掃描(Time-domain OCT)...............9
2.4頻域光學同調斷層掃描(Frequency-domain OCT)..........11
2.5 OCT中常見的假影雜訊................................15
第三章Pix2Pix GAN背景與原理介紹........................16
3.1 Pix2Pix GAN介紹..................................16
3.2生成器架構........................................17
3.3鑑別器架構........................................19
3.4卷積層架構........................................20
3.5圖像評估計算.......................................21
第四章 實驗內容與研究結果..............................23
4.1實驗儀器..........................................23
4.2鏡像雜訊實驗架構....................................24
4.3鏡像雜訊實驗結果....................................28
4.4移動假影實驗架構....................................33
4.5移動假影實驗結果....................................37
第五章 結論...........................................42
參考資料 .............................................43
Extended Abstract....................................45

[1] D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, “Optical coherence tomography,” Science, vol.254, pp. 1178-1181, 1991.
[2] Flournoy PA, McClure RW, Wyntjes G. "White-light interferometric thickness gauge. " Appl Opt. 1972 Sep 1;11(9):1907-15.
[3] Fujimoto JG, Pitris C, Boppart SA, Brezinski ME. "Optical coherence tomography: an emerging technology for biomedical imaging and optical biopsy. " Neoplasia. 2000 Jan-Apr;2(1-2):9-25.
[4] Asrani S, Essaid L, Alder BD, Santiago-Turla C."Artifacts in spectral-domain optical coherence tomography measurements in glaucoma. " JAMA Ophthalmol. 2014 Apr 1;132(4):396-402.
[5] C.M. Huang, E. Wijanto and H.C. Cheng , “Applying a Pix2Pix Generative Adversarial Network to a Fourier-Domain Optical Coherence Tomography System for Artifact Elimination,” IEEE Access, vol. 9, pp. 103311-103324, 2021
[6] Y. Zhang et al., "Neural network-based image reconstruction in swept-source optical coherence tomography using undersampled spectral data," Light: Science & Applications, vol. 10, no. 1, p. 155, 2021.
[7] N. Akter, S. Perry, J. Fletcher, M. Simunovic and M. Roy, "Automated Artifacts and Noise Removal from Optical Coherence Tomography Images Using Deep Learning Technique," 2020 IEEE Symposium Series on Computational Intelligence (SSCI), Canberra, ACT, Australia, 2020, pp. 2536-2542.
[8] A. F. Fercher, C. K. Hitzenberger, G. Kamp, and S. Y. El-Zaiat, "Measurement of intraocular distances by backscattering spectral interferometry," Optics Communications, vol. 117, no. 1, pp. 43-48, 1995/05/15/ 1995.
[9] Chhablani J, Krishnan T, Sethi V, Kozak I. “Artifacts in optical coherence tomography.” Saudi J Ophthalmol. 2014 Apr;28(2):81-7.
[10] Peng Xiao, Viacheslav Mazlin, Kate Grieve, Jose-Alain Sahel, Mathias Fink, and A. Claude Boccara, "In vivo high-resolution human retinal imaging with wavefront-correctionless full-field OCT," Optica 5, 409-412 (2018)
[11] P. Isola, J. Y. Zhu, T. Zhou, and A. A. Efros, ‘‘Image-to-image translation with conditional adversarial networks,’’ IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 5967-5976, 2017.
[12] A. Mino, and G. Spanakis, “LoGAN: Generating Logos with a Generative Adversarial Neural Network Conditioned on color,” IEEE International Conference on Machine Learning and Applications (ICMLA), pp. 965-970, 2018.
[13] O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,” MICCAI, vol. 9351, 2015.
[14] Q. Hao, K. Zhou, J. Yang, Y. Hu, Z. Chai, Y. Ma, G. Liu, Y. Zhao, S. Gao, and J. Liu, ‘‘High signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning,’’ J. Biomed. Opt., vol. 25, no. 12, 2020.
[15] Ugur Demir, Gozde B.Unal, “Patch-Based Image Inpainting with Generative Adversarial Networks,” ArXiv, vol. 1803, 2018.
[16] 維基百科，結構相似性，取自 https://zh.m.wikipedia.org/zh-tw/%E7%B5%90%E6%A7%8B%E7%9B%B8%E4%BC %BC%E6%80%A7
[17] 維基百科，峰值訊噪比，取自 https://zh.wikipedia.org/zh-tw/%E5%B3%B0%E5%80%BC%E4%BF%A1%E5%99%AA%E6%AF%94
[18] By W. Brown PhD, K. Chu PhD, Designed by S. Whitney © Copyright. Lumedica, Inc. 2018. May be used for educational purposes without written permission but with a citation to this source
[19] 廖柏閎, "使用生成對抗網路技術於頻域光學同調斷層掃描圖像加強之研究,"碩士,光電工程系光電與材料科技碩士班,國立虎尾科技大學,雲林縣, 2022.
[20] Parisa Ghaderi Daneshmand, Hossein Rabbani, "Total variation regularized tensor ring decomposition for OCT image denoising and super-resolution",Computers in Biology and Medicine,Volume 177,2024,108591.
[21] R. J. Zawadzki et al., "Progress on Developing Adaptive Optics–Optical Coherence Tomography for In Vivo Retinal Imaging: Monitoring and Correction of Eye Motion Artifacts," in IEEE Journal of Selected Topics in Quantum Electronics, vol. 20, no. 2, pp. 322-333, March-April 2014, Art no. 7100912.
[22] Junchi Liu, Mehmet Kocak, Mark Supanich, Jie Deng, " Motion artifacts reduction in brain MRI by means of a deep residual network with densely connected multi-resolution blocks (DRN-DCMB) " ,Magnetic Resonance Imaging,Volume 71,2020.
[23] Sommer K, Saalbach A, Brosch T, Hall C, Cross NM, Andre JB. "Correction of Motion Artifacts Using a Multiscale Fully Convolutional Neural Network. " AJNR Am J Neuroradiol. 2020 Mar;41(3):416-423.
[24] Bernhard Baumann, Michael Pircher, Erich Götzinger, and Christoph K. Hitzenberger, "Full range complex spectral domain optical coherence tomography without additional phase shifters," Opt. Express 15.