基于同轴相衬成像方法与支持向量机算法的早期骨性关节炎的鉴别诊断

doi:10.3969/j.issn.1006-7795.2022.01.020

首都医科大学学报 ›› 2022, Vol. 43 ›› Issue (1): 120-126.doi: 10.3969/j.issn.1006-7795.2022.01.020

基于同轴相衬成像方法与支持向量机算法的早期骨性关节炎的鉴别诊断

李君^1△, 程程^1△, 董立男², 吴明树³, 张璐^3*

1.北京大学第三医院肿瘤放疗科,北京 100191;
2.中国人民解放军总医院影像科,北京 100039;
3.首都医科大学生物医学工程学院,北京 100069

收稿日期:2021-07-06 出版日期:2022-02-21 发布日期:2022-01-27
作者简介:^△共同第一作者
基金资助:
国家自然科学基金(81401549)。

Differential diagnosis of early osteoarthritis based on in-line phase contrast imaging and support vector machine

Li Jun^1△, Cheng Cheng^1△, Dong Linan², Wu Mingshu³, Zhang Lu^3*

1. Department of Radiation Oncology, Peking University Third Hospital, Beijing 100191, China;
2. Department of Imaging, The General Hospital of the People's Liberation Army, Beijing 100039, China;
3. School of Biomedical Engineering, Capital Medical University, Beijing 100069, China

Received:2021-07-06 Online:2022-02-21 Published:2022-01-27
Contact: * E-mail:luzhang1210@ccmu.edu.cn
Supported by:
National Natural Science Foundation of China (81401549).

摘要/Abstract

摘要： 目的利用同轴相衬成像(in-line phase contrast imaging, IL-PCI)技术与支持向量机(support vector machine, SVM)算法对正常与早期骨性关节炎(osteoarthritis,OA)软骨组织建立分类模型。方法研究样本分别来自接受人工膝关节置换手术及创伤性关节损伤患者的离体膝关节软骨组织。实验对象分为正常组与早期OA组,每组18例。利用IL-PCI技术分别获取两组样本的电子计算机断层扫描(computed tomography, CT)图像,运用行程长度矩阵、分形维度及小波变换三种分析方法对两组样本的CT图像提取纹理参数26个,经主成分分析法降维,利用纹理参数构建支持向量机分类诊断模型,对正常软骨与早期OA软骨进行自动分类。结果利用IL-PCI技术可以清晰观察到正常与早期OA软骨中微结构的变化。两组软骨的纹理特征参数比较显示,大部分纹理特征参数差异有统计学意义。利用SVM算法建立的分类模型能较好地区分正常与早期OA软骨组织,分类准确性可达86.1%。结论利用支持向量机算法与同轴相衬成像方法,能够为早期OA的鉴别诊断提供一种量化判别的手段。

关键词: 同轴相衬成像, 支持向量机, 骨关节炎, 纹理特征

Abstract: Objective To establish the classification model for normal and early osteoarthritis (OA) of cartilage tissue by in-line phase contrast imaging (IL-PCI) technique and support vector machine (SVM) algorithm. Methods The study samples were in vitro knee cartilage samples from patients undergoing arthroplasty. The research subjects were divided into normal group and early OA group, 18 cases in each group. Computed tomography (CT) images of cartilage tissues were obtained by in-line phase contrast imaging (IL-PCI) technology. Totally 26 texture parameters were extracted from the two groups of CT images by using three analysis methods of stroke length matrix, fractal dimension and discrete wavelet transform. Dimensionality of texture parameters was reduced by principal component analysis. A SVM classification model was established based on texture parameters to classify normal cartilage and early OA cartilage automatically. Results The changes of microstructure in normal and early OA cartilage could be clearly observed by IL-PCI technique. The cartilage texture parameters of the two groups were compared, and the results showed that most of the texture parameters were significantly different. The classification accuracy of the two groups can reach 86.1% by support vector machine algorithm. Conclusion The support vector machine algorithm and IL-PCI method can provide an effective auxiliary method for the diagnosis of early osteoarthritis.

Key words: in-line phase contrast imaging (IL-PCI), support vector machine, osteoarthritis, texture feature

中图分类号:

R684.3

李君, 程程, 董立男, 吴明树, 张璐. 基于同轴相衬成像方法与支持向量机算法的早期骨性关节炎的鉴别诊断[J]. 首都医科大学学报, 2022, 43(1): 120-126.

Li Jun, Cheng Cheng, Dong Linan, Wu Mingshu, Zhang Lu. Differential diagnosis of early osteoarthritis based on in-line phase contrast imaging and support vector machine[J]. Journal of Capital Medical University, 2022, 43(1): 120-126.

参考文献

[1] 任鹏玲,牛海军,宫赫,等. 老龄关节软骨和软骨下骨的变化及相互作用机制研究进展[J]. 医用生物力学, 2020, 35(2): 253-258.
[2] 常亮, 秦江辉, 史冬泉, 等. 骨关节炎与软骨下骨研究进展[J]. 中华骨与关节外科杂志, 2019, 12(10): 827-832.
[3] 姜丽萍,陆皓,汉瑞娟,等.老年膝关节骨性关节炎患者跌倒风险临床预测规则的构建[J]. 中国医科大学学报,2020,49(6):541-545
[4] Monnin P, Hlushchuk R, Djonov V, et al. X-ray phase contrast imaging of metallized and non-metallized biological samples[J]. Z Med Phys, 2006, 16(3): 172-178.
[5] Andrews J C, Meirer F, Liu Y J, et al. Transmission X-ray microscopy for full-field nano imaging of biomaterials[J]. Microsc Res Tech, 2011, 74(7): 671-681.
[6] Barbone G E, Bravin A, Romanelli P, et al. Micro-imaging of brain cancer radiation therapy using phase-contrast computed tomography[J]. Int J Radiat Oncol Biol Phys, 2018, 101(4): 965-984.
[7] Duan J H, Hu C H, Qiu Q T, et al. Characterization of microvessels and parenchyma in in-line phase contrast imaging CT: healthy liver, cirrhosis and hepatocellular carcinoma[J]. Quant Imaging Med Surg, 2019, 9(6): 1037-1046.
[8] Han S M, Chikawa J I, Jeon J K, et al. Synchrotron nanoscopy imaging study of scalp hair in breast cancer patients and healthy individuals: difference in medulla loss and cortical membrane enhancements[J]. Microsc Res Tech, 2016, 79(1): 23-30.
[9] Seidler M, Forghani B, Reinhold C, et al. Dual-energy CT texture analysis with machine learning for the evaluation and characterization of cervical lymphadenopathy[J]. Comput Struct Biotechnol J, 2019, 17: 1009-1015.
[10] Andersen M B, Harders S W, Ganeshan B, et al. CT texture analysis can help differentiate between malignant and benign lymph nodes in the mediastinum in patients suspected for lung cancer[J]. Acta radiol, 2016, 57(6): 669-676.
[11] Li J, Yuan H H, Wu M S, et al. Quantitative assessment of murine articular cartilage and bone using X-ray phase-contrast imaging[J]. Plos One, 2014, 9(11): e111939.
[12] 刘丽, 匡纲要. 图像纹理特征提取方法综述[J]. 中国图象图形学报, 2009, 14(4): 622-635.
[13] Hatt M, Tixier F, Pierce L, et al. Characterization of PET/CT images using texture analysis: the past, the present...any future?[J]. Eur J Nucl Med Mol Imaging, 2017, 44(1): 151-165.
[14] 陈明丽, 陈亚青, 朱云开. 利用主成分分析提取人类肝纤维化声像图的纹理特征[J]. 肝脏, 2014, 19(10): 756-758.
[15] 吴饶仙, 况九龙. 应用主成分和聚类分析法探索慢性阻塞性肺疾病表型的临床研究[J]. 中国全科医学, 2014, 17(31): 3699-3703.
[16] Almansour N A, Syed H F, Khayat N R, et al. Neural network and support vector machine for the prediction of chronic kidney disease: a comparative study[J]. Comput Biol Med, 2019, 109: 101-111.
[17] Horng A, Brun E, Mittone A, et al. Cartilage and soft tissue imaging using X-rays: propagation-based phase-contrast computed tomography of the human knee in comparison with clinical imaging techniques and histology[J]. Invest Radiol, 2014, 49(9): 627-634.
[18] Tadimalla S, Tourell M C, Knott R, et al. Assessment of collagen fiber orientation dispersion in articular cartilage by small-angle X-ray scattering and diffusion tensor imaging: preliminary results[J]. Magn Reson Imaging, 2018, 48: 115-121.

基于同轴相衬成像方法与支持向量机算法的早期骨性关节炎的鉴别诊断

Differential diagnosis of early osteoarthritis based on in-line phase contrast imaging and support vector machine

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

本文评价

[1]	王振中, 喻飞, 杨波, 马立峰. 富血小板血浆对兔骨关节炎软骨中基质金属蛋白酶-13表达的影响[J]. 首都医科大学学报, 2020, 41(6): 923-928.
[2]	李守龙, 艾依热提·买买提, 潘海涛, 曾峥. 膝骨关节炎滑膜不同部位基质金属蛋白酶3的表达及临床意义[J]. 首都医科大学学报, 2013, 34(3): 411-415.