MicroCT图像中半月板周向纤维的分割方法与三维重建

HTML

PDF (PC)

MicroCT图像中半月板周向纤维的分割方法与三维重建

Segmentation and 3D Reconstruction of Meniscus Circumferential Fibers in MicroCT Images

MicroCT图像中半月板周向纤维的分割方法与三维重建

Segmentation and 3D Reconstruction of Meniscus Circumferential Fibers in MicroCT Images

摘要/Abstract

参考文献

Metrics

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 0

doi:10.12052/gdutxb.230167

广东工业大学学报 ›› 2025, Vol. 42 ›› Issue (1): 42-50.doi: 10.12052/gdutxb.230167

• 智慧医疗 • 上一篇

王彪¹, 钟映春¹, 罗唯师², 朱爽³, 曾蒲军⁴

1. 广东工业大学自动化学院, 广东广州 510006;
2. 广东省第二人民医院神经外科, 广东广州 510317;
3. 南方医科大学珠江医院关节骨病外科, 广东广州 510280;
4. 湖南省药品审核查验中心, 湖南长沙 410000

收稿日期:2023-10-23 发布日期:2025-01-14
通信作者: 钟映春(1973–) ，男，副教授，博士，主要研究方向为神经信息学与图像理解，E-mail：gdut_zyc@qq.com
作者简介:王彪(2000–) ，男，硕士研究生，主要研究方向为模式识别与图像处理，E-mail：1769517568@qq.com
基金资助:
国家自然科学基金资助项目(61975248)

Wang Biao¹, Zhong Yingchun¹, Luo Weishi², Zhu Shuang³, Zeng Pujun⁴

1. School of Automation, Guangdong University of Technology, Guangzhou 510006, China;
2. Department of Neurosurgery, Second People's Hospital of Guangdong Province, Guangzhou 510317, China;
3. Department of Arthropathy, Zhujiang Hospital, Southern Medical University, Guangzhou 510280, China;
4. Hunan Drug Inspection Center, Changsha 410000, China

Received:2023-10-23 Published:2025-01-14

摘要： 周向纤维是半月板受力的关键区域。构建周向纤维的三维微观结构对于半月板损伤的治疗和人工半月板的研发具有重要意义。当前主要通过人工对微计算机断层扫描技术(Micro Computed Tomography, MicroCT)图像中半月板的周向纤维进行分割。由于半月板微观组织复杂，人工分割存在效率低下、分割标准不一致等问题。本文针对样本图像较少的问题，结合MicroCT图像的特点提出了一种图像扩增方法；针对边缘分割误差大的问题，提出了一种基于TransUNet算法的改进模型，引入了图像块相对位置编码(image Relative Position Encoding, iRPE)并改进了损失函数。实验结果表明：(1) 改进模型能精确完整分割半月板组织，分割结果可成功完成周向纤维三维重建。(2) 引入的iRPE算法提高了模型边缘细节的分割效果，改进的损失函数使模型更好地适应样本不均衡情况，提出的图像扩增方法解决了数据集不足的问题，综合提升了模型的性能；结果显示周向纤维分割的平均精度达到98.66%。(3) 在周向纤维三维模型中发现纤维发生分裂的现象，且以一分为二为主，同时存在少量一分为三的情况。本文提出的方法能高精度和高效率分割MicroCT图像中半月板周向纤维，为研究半月板在三维空间的受力分析做好铺垫。

关键词: 半月板, 周向纤维, 三维重建, MicroCT图像, 深度学习

Abstract: The circumferential fibers are the key areas of meniscus stress. The construction of three-dimensional microstructure of circumferential fibers is of great significance for the treatment of meniscus injury and the development of artificial meniscus. At present, the circumferential fibers of meniscus in micro computed tomography (MicroCT) images are segmented manually. Because of the complex microstructure of meniscus, manual segmentation has some problems such as low efficiency and inconsistent segmentation standards. To solve the problem of few sample images, an image amplification method is proposed based on the characteristics of MicroCT images. To solve the problem of large edge segmentation errors, an improved model based on TransUNet algorithm is proposed, image Relative Position Encoding (iRPE) is introduced, and the loss function is improved. The experimental results show that: (1) the improved model can accurately and completely segment the meniscus tissue, and the segmentation results can successfully complete the three-dimensional reconstruction of circumferential fibers. (2) The introduced iRPE algorithm improves the segmentation effect of model edge details, the improved loss function enables the model to better adapt to the situation of sample imbalance, and the proposed image amplification method solves the problem of insufficient data sets and comprehensively improves the performance of the model.The results show that the average precision of circumferential fiber segmentation is 98.66%. (3) In the three-dimensional model of circumferential fibers, it is found that fibers are divided into two parts, and a small amount of fibers are divided into three parts. The proposed method can segment the meniscus circumferential fibers in MicroCT images with high accuracy and efficiency, and can pave the way for the study of the force analysis of meniscus in three-dimensional space.

Key words: meniscus, circumferential fibers, three-dimensional reconstruction, MicroCT images, deep learning

中图分类号:

R318，TP391

王彪, 钟映春, 罗唯师, 朱爽, 曾蒲军. MicroCT图像中半月板周向纤维的分割方法与三维重建[J]. 广东工业大学学报, 2025, 42(1): 42-50.doi: 10.12052/gdutxb.230167

Wang Biao, Zhong Yingchun, Luo Weishi, Zhu Shuang, Zeng Pujun. Segmentation and 3D Reconstruction of Meniscus Circumferential Fibers in MicroCT Images[J]. Journal of Guangdong University of Technology, 2025, 42(1): 42-50.doi: 10.12052/gdutxb.230167

[1] ZIHNA G, TOPUZ B, GÜNAL G, et al. Preparation of hybrid meniscal constructs using hydrogels and acellular matrices [J]. Journal of Biomaterials Science, Polymer Edition, 2023, 34(5): 587-611.
[2] PRENDERGAST M E, HEO S J, MAUCK R L, et al. Suspension bath bioprinting and maturation of anisotropic meniscal constructs [J]. Biofabrication, 2023, 15(3): 035003.
[3] BILGEN B, JAYASURIYA C T, OWENS B D. Current concepts in meniscus tissue engineering and repair [J]. Advanced Healthcare Materials, 2018, 7(11): 1701407.
[4] 朱爽, 梁敏娟, 赖剑强. 基于Micro-CT技术的半月板微观结构分析和三维重建[J]. 现代医用影像学, 2021, 30(8): 1432-1434.
[5] KARJALAINEN V P, KESTILÄ I, FINNILÄ M A, et al. Quantitative three-dimensional collagen orientation analysis of human meniscus posterior horn in health and osteoarthritis using micro-computed tomography [J]. Osteoarthritis and Cartilage, 2021, 29(5): 762-772.
[6] PETERSEN W, TILLMANN B. Collagenous fibril texture of the human knee joint menisci [J]. Anatomy and Embryology, 1998, 197: 317-324.
[7] FOX A J S, BEDI A, RODEO S A. The basic science of human knee menisci: structure, composition, and function [J]. Sports Health, 2012, 4(4): 340-351.
[8] TACK A, MUKHOPADHYAY A, ZACHOW S. Knee menisci segmentation using convolutional neural networks: data from the osteoarthritis initiative [J]. Osteoarthritis and Cartilage, 2018, 26(5): 680-688.
[9] SHELHAMER E, LONG J, DARRELL T. Fully convolutional networks for semantic segmentation [J]. IEEE Trans Pattern Anal Mach Intell, 2017, 39(4): 640-651.
[10] RONNEBERGER O, FISCHER P, BROX T. U-net: convolutional networks for biomedical image segmentation[EB/OL]. arXiv: 1505.04597 (2015-05-18) [2023-09-15]. https://doi.org/10.48550/arXiv.1505.04597.
[11] STEGMANN H, WERKMEISTER R M, PFISTER M, et al. Deep learning segmentation for optical coherence tomography measurements of the lower tear meniscus [J]. Biomedical Optics Express, 2020, 11(3): 1539-1554.
[12] CHEN J H, LU Y Y, YU Q H, et al. TransUNet: Transformers make strong encoders for medical image segmentation[EB/OL]. arXiv: 2102.04346 (2021-02-08) [2023-09-15]. https://doi.org/10.48550/arXiv.2102.04306.
[13] FEDOROV A, BEICHEL R, KALPATHY-CRAMER J, et al. 3D Slicer as an image computing platform for the quantitative imaging network [J]. Magnetic Resonance Imaging, 2012, 30(9): 1323-1341.
[14] ZHU S, GE T, XIANG J P, et al. Microstructure analysis and reconstruction of a meniscus [J]. Orthopaedic Surgery, 2021, 13(1): 306-313.
[15] RUSSELL B C, TORRALBA A, MURPHT K P, et al. LabelMe: a database and web-based tool for image annotation [J]. International Journal of Computer Vision, 2008, 77: 157-173.
[16] WOODMASS J M, LAPRADE R F, SGAGLIONE N A, et al. Meniscal repair: reconsidering indications, techniques, and biologic augmentation [J]. The Journal of Bone and Joint surgery. American Volume, 2017, 99(14): 1222-1231.SS.
[17] WU K, PENG H, CHEN M, et al. Rethinking and improving relative position encoding for vision transformer[EB/OL]. arXiv: 2107.14222 (2021-07-29) [2023-09-15]. https://doi.org/10.48550/arXiv.2107.14222.
[18] SHAW P, USZKOREIT J, VASWANI A. Self-attention with relative position representations[EB/OL]. arXiv: 1803.02155 (2018-04-12) [2023-09-15]. https://doi.org/10.48550/arXiv.1803.02155.
[19] MILLETARI F, NAVAB N, AHMADI S A. V-Net: fully convolutional neural networks for volumetric medical image segmentation[EB/OL]. arXiv: 1606.04797 (2016-06-15) [2023-09-15]. https://doi.org/10.48550/arXiv.1606.04797.
[20] ROTH H R, ODA H, ZHOU X, et al. An application of cascaded 3D fully convolutional networks for medical image segmentation [J]. Computerized Medical Imaging and Graphics, 2018, 66: 90-99.
[21] 孙思语, 钟映春, 郑海阳, 等. MicroCT图像中周围神经束轮廓的建模[J]. 计算机仿真, 2022, 39(1): 258-262.
SUN S Y, ZHONG Y C, ZHENG H Y, et al. Modeling of contour of peripheral fasciculi in MicroCT images [J]. Computer Simulation, 2022, 39(1): 258-262.
[22] GOMALAVALLI R, MUTTAN S, SAI P M V. Boundary detection of renal using contour segmentation [J]. International Journal of Biomedical Engineering and Technology, 2018, 28(1): 53-66.

Viewed

Full text

From	Others	local

Times	13	68
Rate	16%	84%

Abstract

Just accepted	Online first	Issue

0	10	76

	From	local

	Times	86
	Rate	100%

Cited

Web of Science	Crossref	ScienceDirect	Search for Citations in Google Scholar >>


This page requires you have already subscribed to WoS.

Shared

Discussed

Just accepted

Online first

Just accepted

Online first

[1]	王盈, 蔡述庭, 熊晓明. 基于主动学习和视觉状态空间模型的热点检测器[J]. 广东工业大学学报, 2024, 41(06): 45-51.
[2]	罗成, 张军. 基于深度学习的自适应采样及记忆增强压缩感知方法[J]. 广东工业大学学报, 2024, 41(04): 114-121.
[3]	梁宇辰, 蔡念, 欧阳文生, 谢依颖, 王平. 基于切片关联信息的慢性阻塞性肺疾病CT诊断[J]. 广东工业大学学报, 2024, 41(01): 27-33.
[4]	吴菊华, 李俊锋, 陶雷. 基于知识图谱嵌入与深度学习的药物不良反应预测[J]. 广东工业大学学报, 2024, 41(01): 19-26,40.
[5]	温雯, 刘莹, 蔡瑞初, 郝志峰. 面向多粒度交通流预测的时空深度回归模型[J]. 广东工业大学学报, 2023, 40(04): 1-8.
[6]	钟耿君, 李东. 基于通道分离机制的双分支点云处理网络[J]. 广东工业大学学报, 2023, 40(04): 18-23.
[7]	金宇凯, 李志生, 欧耀春, 张华刚, 曾江毅, 陈搏超. 基于多阶段聚类的PM_2.5质量浓度预测及对比研究[J]. 广东工业大学学报, 2023, 40(03): 17-24.
[8]	吴俊贤, 何元烈. 基于通道注意力的自监督深度估计方法[J]. 广东工业大学学报, 2023, 40(02): 22-29.
[9]	刘冬宁, 王子奇, 曾艳姣, 文福燕, 王洋. 基于复合编码特征LSTM的基因甲基化位点预测方法[J]. 广东工业大学学报, 2023, 40(01): 1-9.
[10]	徐伟锋, 蔡述庭, 熊晓明. 基于深度特征的单目视觉惯导里程计[J]. 广东工业大学学报, 2023, 40(01): 56-60,76.
[11]	刘洪伟, 林伟振, 温展明, 陈燕君, 易闽琦. 基于MABM的消费者情感倾向识别模型——以电影评论为例[J]. 广东工业大学学报, 2022, 39(06): 1-9.
[12]	章云, 王晓东. 基于受限样本的深度学习综述与思考[J]. 广东工业大学学报, 2022, 39(05): 1-8.
[13]	郑佳碧, 杨振国, 刘文印. 基于细粒度混杂平衡的营销效果评估方法[J]. 广东工业大学学报, 2022, 39(02): 55-61.
[14]	Gary Yen, 栗波, 谢胜利. 地球流体动力学模型恢复的长短期记忆网络渐进优化方法[J]. 广东工业大学学报, 2021, 38(06): 1-8.
[15]	赖峻, 刘震宇, 刘圣海. 基于全局数据混洗的小样本数据预测方法[J]. 广东工业大学学报, 2021, 38(03): 17-21.