搜索

x

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于机器学习的扫频OCT成像畸变校正

马翠 陈沛哲 杨璐 韩涛 汤赟 陈昌勇 丁志华

引用本文:
Citation:

基于机器学习的扫频OCT成像畸变校正

马翠, 陈沛哲, 杨璐, 韩涛, 汤赟, 陈昌勇, 丁志华

A Machine Learning Approach to Correct Imaging Distortions in Swept-Source OCT

MA Cui, CHEN Peizhe, YANG Lu, HAN Tao, TANG Yun, CHEN Changyong, DING Zhihua
Article Text (iFLYTEK Translation)
PDF
导出引用
  • 光学相干断层扫描(Optical Coherence Tomography,简记为OCT)成像中不可避免的畸变常常会导致成像空间与真实空间之间的不匹配,影响测量的准确性。为解决这一问题,本研究提出了一种基于机器学习的OCT图像畸变校正方法。首先,将带有均匀分布圆孔阵列的校准板依次在不同标记平面进行成像。选取坐标与所有成像平面的平均坐标偏差最小的点作为参考标记点。然后,利用数学模型重构参考平面上所有标记点的坐标,从而建立校准板成像空间与真实物理空间之间的映射关系。采用多层感知机来学习这种映射关系。利用训练完成的模型推导整个空间点的分布规律,从而实现透镜OCT图像的畸变校正,并求出透镜中心厚度与曲率半径。校正后透镜曲率半径精度达到10μm,误差在1%以内。中心厚度精度可达3μm,相对误差为0.3%。
    The inevitable distortions in optical coherence tomography (OCT) imaging often lead to mismatches between the imaging space and the real space, significantly affecting measurement accuracy. To address this issue, this study proposes a machine learning-based OCT image distortion correction method. A calibration plate with uniformly distributed circular hole arrays was sequentially imaged at different marked planes. The point showing minimal deviation between its coordinates and the mean coordinates across all imaging planes was selected as the reference marker. A mathematical model was then used to reconstruct all marker point coordinates in the reference plane, establishing a mapping relationship between the calibration plate's imaging space and the real physical space. A multilayer perceptron (MLP) was employed to learn this mapping relationship. The network architecture consisted of multiple fully connected modules, each containing a linear layer and an activation function except for the output layer. The optimal model was selected based on validation set performance and subsequently applied to analyze the spatial distribution of points. Using a swept-source OCT system, lens images were acquired and corrected through the trained model to obtain the anterior surface point cloud. Combined with ray tracing reconstruction of the posterior surface, the lens curvature radius and central thickness were calculated. Experimental results demonstrated that after correction, the lens curvature radius was measured with an accuracy of 10μm (error < 1%), while the central thickness was determined with a precision of 3μm (relative error: 0.3%). This method demonstrates high precision and reliability, offering an effective solution for improving OCT measurement accuracy.
  • [1]

    Huang D, Swanson E A, Lin C P, Schuman J S, Stinson W G, Chang W, Hee M R, Flotte T, Gregory K, Puliafito C A, Fujimoto J G 1991Science 254 1178

    [2]

    Tomlins P H, Wang R K 2005J. Phys. D- Appl. Phys. 38 2519

    [3]

    Ding Y T, Zhang J, Guo Y, Chen H 2024Laser Optoelectron. Prog. 61 247(in Chinese) [丁宇韬,张君,郭遥,陈昊2024激光与光电子学进展61 247]

    [4]

    Volker W, Andrew R, Sunita R, Joseph I 2002Opt. Express 10 397

    [5]

    Diaz J, Rahlves M, Majdani O, Reithmeier E, Ortmaier T 2012Optical Metrology and Inspection for Industrial Applications II Beijing, China, November 5-7,2012 p8563J

    [6]

    Yao J, Anderson A, Rolland J P 2018Opt. Express 26 10242

    [7]

    Jiang X Q, Fan Y Q, Wang W 2014J. Front. Comput. Sci. Technol 8 1254(in Chinese) [江祥奎, 范永青, 王婉2014计算机科学与探索8 1254]

    [8]

    Qi Z S, Wang Z, Huang J H, Xue Q, Gao J M 2016Acta Photonica Sin. 45 87(in Chinese) [齐召帅,王昭,黄军辉,薛琦,高建民2016光子学报45 87]

    [9]

    Lecun Y, Bottou L, Bengio Y, Haffner P 1998Proc. IEEE 86 2278

    [10]

    Li X Y, Zhang B, Sander P V, Liao J 2019Computer Vision and Pattern Recognition Long Beach, CA,USA, June 15-20,2019 p4850

    [11]

    Yang S R, Lin C Y, Liao K, Zhang C J, Zhao Y 2021 arXiv:2103.16026v2[cs.CV]

    [12]

    Xu Q W, Wang P P, Zeng Z J, Huang Z B, Zhou X X, Liu J M, Li Y, Chen S Q, Fan D Y 2020Acta Phys. Sin. 69 014209(in Chinese) [徐启伟, 王佩佩, 曾镇佳, 黄泽斌, 周新星, 刘俊敏, 李瑛, 陈书青, 范滇元2020物理学报69 014209]

    [13]

    Zhou J, Zhang X F, Zhao Y G 2021Acta Phys. Sin. 70 054201(in Chinese) [周静, 张晓芳, 赵延庚2021物理学报70 054201]

    [14]

    Shukla S, Vishwakarma C, Sah A N, Ahirwar S, Pandey K, Pradhan A 2023Appl. Opt. 62 6826

    [15]

    Breiman L, 2001Mach. Learn. 45 5

    [16]

    Xie T Y, Li Y M, Zhang T, Gao T L, Shi Y C 2023Mod. Mach. Tool Autom. Manuf. Tech. 237(in Chinese) [谢探阳,李玉梅,张涛,高天亮,石玉超2023组合机床与自动化加工技术237]

    [17]

    Kato D, Maeda N, Hirogaki T, Aoyama E, Takahashi K 202121st International Conference on Control, Automation and Systems Jeju, Korea, October 12-15,2021 p607

    [18]

    Werbos P J 1981System Modeling and Optimization Berlin, Germany, July 20-24, 1981 p762

    [19]

    Rumelhart D E, Hinton G E, Williams R J 1986Nature 323 533

    [20]

    Wang Y L, Zhang N, Zhang X Y, Xu Y Z, Wang M 2025Acta Opt. Sin. 45 3(in Chinese) [王艳丽,张宁,张祥宇,许益彰,王密2025光学学报45 3]

    [21]

    Singarimbun R N, Nababan E B, Sitompul O S 2019International Conference of Computer Science and Information Technology Medan, Indonesia, November 28-29,2019 p1

    [22]

    Cotter A, Shamir O, Srebro N, Sridharan K 2011Advances in Neural Information Processing Systems Granada, Spain, December 12-15, 2011 p1647

    [23]

    Fitzgibbon A, Pilu M, Fisher R B 1999IEEE Trans. Pattern Anal. Mach. Intell. 21 476

  • [1] 高梦琴, 章慧英, 孙彬, 刘开元, 杜持新, 李鹏. 基于光学相干血流造影引导的近红外一区视网膜血氧饱和度测量. 物理学报, doi: 10.7498/aps.74.20241686
    [2] 郭焱, 吕恒, 丁春玲, 袁晨智, 金锐博. 分数阶涡旋光衍射过程的机器学习识别. 物理学报, doi: 10.7498/aps.74.20241458
    [3] 张童, 王加豪, 田帅, 孙旭冉, 李日. 基于机器学习的铸件凝固过程动态收缩行为. 物理学报, doi: 10.7498/aps.74.20241581
    [4] 王鹏, 麦麦提尼亚孜·麦麦提阿卜杜拉. 机器学习的量子动力学. 物理学报, doi: 10.7498/aps.74.20240999
    [5] 张旭, 丁进敏, 侯晨阳, 赵一鸣, 刘鸿维, 梁生. 基于机器学习的激光匀光整形方法. 物理学报, doi: 10.7498/aps.73.20240747
    [6] 张嘉晖. 蛋白质计算中的机器学习. 物理学报, doi: 10.7498/aps.73.20231618
    [7] 郭唯琛, 艾保全, 贺亮. 机器学习回归不确定性揭示自驱动活性粒子的群集相变. 物理学报, doi: 10.7498/aps.72.20230896
    [8] 刘烨, 牛赫然, 李兵兵, 马欣华, 崔树旺. 机器学习在宇宙线粒子鉴别中的应用. 物理学报, doi: 10.7498/aps.72.20230334
    [9] 管星悦, 黄恒焱, 彭华祺, 刘彦航, 李文飞, 王炜. 生物分子模拟中的机器学习方法. 物理学报, doi: 10.7498/aps.72.20231624
    [10] 应大卫, 张思慧, 邓书金, 武海斌. 基于机器学习的单拍冷原子成像. 物理学报, doi: 10.7498/aps.72.20230449
    [11] 张嘉伟, 姚鸿博, 张远征, 蒋伟博, 吴永辉, 张亚菊, 敖天勇, 郑海务. 通过机器学习实现基于摩擦纳米发电机的自驱动智能传感及其应用. 物理学报, doi: 10.7498/aps.71.20211632
    [12] 林键, 叶梦, 朱家纬, 李晓鹏. 机器学习辅助绝热量子算法设计. 物理学报, doi: 10.7498/aps.70.20210831
    [13] 陈江芷, 杨晨温, 任捷. 基于波动与扩散物理系统的机器学习. 物理学报, doi: 10.7498/aps.70.20210879
    [14] 吴彤, 霍文麒, 黄蕴智, 王吉明, 顾晓蓉, 路元刚, 赫崇君, 刘友文. 用于内窥光学相干层析成像的小型化预标定Lissajous扫描光纤探头. 物理学报, doi: 10.7498/aps.70.20210151
    [15] 张瑶, 张云波, 陈立. 基于深度学习的光学表面杂质检测. 物理学报, doi: 10.7498/aps.70.20210403
    [16] 杨自欣, 高章然, 孙晓帆, 蔡宏灵, 张凤鸣, 吴小山. 铅基钙钛矿铁电晶体高临界转变温度的机器学习研究. 物理学报, doi: 10.7498/aps.68.20190942
    [17] 胡喆皓, 上官紫微, 邱建榕, 杨珊珊, 鲍文, 沈毅, 李鹏, 丁志华. 基于受激辐射信号的谱域光学相干层析分子成像方法. 物理学报, doi: 10.7498/aps.67.20171738
    [18] 唐弢, 赵晨, 陈志彦, 李鹏, 丁志华. 超高分辨光学相干层析成像技术与材料检测应用. 物理学报, doi: 10.7498/aps.64.174201
    [19] 赵晨, 陈志彦, 丁志华, 李鹏, 沈毅, 倪秧. 线照明并行谱域光学相干层析成像系统与缺陷检测应用研究. 物理学报, doi: 10.7498/aps.63.194201
    [20] 商在明, 丁志华, 王玲, 刘勇. 基于光程编码与相干合成的三维超分辨术. 物理学报, doi: 10.7498/aps.60.124204
计量
  • 文章访问数:  17
  • PDF下载量:  0
  • 被引次数: 0
出版历程
  • 上网日期:  2025-07-01

/

返回文章
返回