融合先验知识特征的超声图像甲状旁腺结节识别

doi:10.12263/DZXB.20200271

电子学报 ›› 2021, Vol. 49 ›› Issue (5): 944-952.DOI: 10.12263/DZXB.20200271

融合先验知识特征的超声图像甲状旁腺结节识别

毛林¹, 赵利强¹, 于明安², 魏莹², 王颖^3,4

1. 北京化工大学信息科学与技术学院, 北京 100029;
2. 中日友好医院介入超声医学科, 北京 100029;
3. 中国科学院微电子研究所, 北京 100029;
4. 中国科学院大学, 北京 100049

收稿日期:2020-03-17 修回日期:2020-06-08 出版日期:2021-05-25
- 作者简介:
- 毛林女,1995年出生,湖北随州人.现为北京化工大学信息科学与技术学院硕士研究生,研究方向为基于超声图像的甲状旁腺结节识别方法. E-mail:1244070607@qq.com;赵利强 男,1982年出生,河南洛阳人.现为北京化工大学信息科学与技术学院副教授,研究方向为智能检测技术. E-mail:zhaolq@mail.buct.edu.cn;王颖女,1969年出生,天津人.现为中国科学院微电子研究所副研究员,研究方向为机器视觉检测和图像分析. E-mail:wangying@mail.buct.edu.cn
- 基金资助:
- 北京化工大学-中日友好医院生物医学转化工程联合基金 (No.PYBZ1804）

Recognition of Parathyroid Nodule by Fusing Prior Knowledge Features in Ultrasound Image

MAO Lin¹, ZHAO Li-qiang¹, YU Ming-an², WEI Ying², WANG Ying^3,4

1. College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China;
2. Interventional Ultrasound Medicine, China-Japan Friendship Hospital, Beijing 100029, China;
3. Institute of Microelectronics of Chinese Academy of Sciences, Beijing 100029, China;
4. University of Chinese Academy of Sciences, Beijing 100049, China

Received:2020-03-17 Revised:2020-06-08 Online:2021-05-25 Published:2021-05-25
- Supported by:
- Beijing University of Chemical Technology - China-Japan Friendship Hospital Biomedical Translational Engineering Research Center Joint Fund (No.PYBZ1804)

摘要/Abstract

摘要： 正确识别超声图像中的甲状旁腺结节对甲状旁腺功能亢进的诊断治疗非常重要.由于病人个体的差异性和超声图像的复杂性,采用图像的形态特征和纹理特征识别甲状旁腺结节准确率低.本文提出利用包膜以及结节与甲状腺相对位置的先验知识特征描述方法,并将其与形态、纹理特征融合,采用支持向量数据描述(Support Vector Data Description,SVDD)识别甲状旁腺结节.实验结果表明,先验知识特征可以准确描述甲状旁腺结节的特征,融合先验知识特征比仅利用形态特征和纹理特征具有更高的识别准确率.

关键词: 甲状旁腺结节, 先验知识特征, SVDD, 超声图像

Abstract: It is very important to recognize parathyroid nodules correctly in ultrasound images for the treatment of hyperparathyroidism. Due to individual differences of patients and complexity of ultrasound images, parathyroid nodules can’t be recognized accurately by only using morphological features and texture features. In this paper, a prior knowledge feature description method is proposed on account of the characteristic of envelope and the relative location between the nodule and the thyroid. SVDD is applied to recognize parathyroid nodules based on the fusion features of prior knowledge features, morphological features and texture features. The experimental results show that the prior knowledge features can describe the characteristics of parathyroid nodules well, and the accuracy by using the fusion features which combined prior knowledge features is higher than that of only using morphological features and texture features for the recognition of parathyroid nodules.

Key words: parathyroid nodules, prior knowledge features, SVDD, ultrasound images

中图分类号:

TP391.4

毛林, 赵利强, 于明安, 等. 融合先验知识特征的超声图像甲状旁腺结节识别[J]. 电子学报, 2021, 49(5): 944-952.

MAO Lin, ZHAO Li-qiang, YU Ming-an, et al. Recognition of Parathyroid Nodule by Fusing Prior Knowledge Features in Ultrasound Image[J]. Acta Electronica Sinica, 2021, 49(5): 944-952.

参考文献

[1] Hedgeman E,Lipworth L,Lowe K,et al.International burden of chronic kidney disease and secondary hyperparathyroidism:A systematic review of the literature and available data[J].International Journal of Nephrology,2015,2015:1-15.
[2] Johnson N A,Tublin M E,Ogilvie J B.Parathyroid imaging:technique and role in the preoperative evaluation of primary hyperparathyroidism[J].Ajr Am J Roentgenol,2007,188(6):1706-1715.
[3] 李春芳,杨鑫,张旭明,等.基于MAP的超声图像分解去噪算法研究[J].电子学报,2014,42(7):1291-1298. Li C F,Yang X,Zhang X M,et al.MAP based ultrasound image decomposition and denoising method[J].Acta Electronica Sinica,2014,42(7):1291-1298.(in Chinese).
[4] Senchenkov A,Staren E D.Ultrasound in head and neck surgery:Thyroid,parathyroid,and cervical lymph nodes[J].Surgical Clinics of North America,2004,84(4):973-1000.
[5] 林其忠,余建国,赵暖,等.乳腺肿瘤超声图像的特征分析[J].仪器仪表学报,2006(S1):744-746,752. Lin Q Z,Yu J G,Zhao N,et al.Feature analysis of ultrasound breast tumor images[J].Chinese Journal of Scientific Instrument,2006(S1):744-746,752.(in Chinese).
[6] Akkasaligar P T.Classification of medical ultrasound images of kidney[A].IJCA Online Proceedings on International Conference on Information and Communication Technologies[C].Kochi,India:ICICT,2014,3:24-28.
[7] 雷一鸣,赵希梅,于可歆,等.鲁棒的超声图像肝硬化识别方法[J].电子测量与仪器学报,2016,30(10):1460-1467. Lei Y M,Zhao X M,Yu K X,et al.A robust method for cirrhosis recognition in ultrasound image[J].Journal of Electronic Measurement and Instrumentation,2016,30(10):1460-1467.(in Chinese).
[8] 王昕,李亮,尹小童,等.甲状腺结节超声图像多特征融合及识别[J].吉林大学学报(信息科学版),2017,35(06):650-655. Wang X,Li L,Yin X T,et al.Multiple feature fusion and recognition of thyroid nodule ultrasound image[J].Journal of Jilin University(Information Science Edition),2017,35(06):650-655.(in Chinese).
[9] Ardakani A A,Gharbali A,Mohammadi A.Classification of benign and malignant thyroid nodules using wavelet texture analysis of sonograms[J].Journal of Ultrasound in Medicine,2015,34(11):1983-1989.
[10] Acharya U R,Chowriappa P,Fujita H,et al.Thyroid lesion classification in 242 patient population using Gabor transform features from high resolution ultrasound images[J].Knowledge-Based Systems,2016,107:235-245.
[11] Ghervan C.Thyroid and parathyroid ultrasound[J].Medical Ultrasonography,2011,13(1):80-84.
[12] 邵选,潘永海,吴成亮,等.精细被膜解剖法原位保留甲状旁腺的临床应用[J].实用医学杂志,2009,25(13):2110-2111.
[13] 陈金华,刘滨月,贾宝霞,等.甲状旁腺瘤的超声诊断及漏误诊原因分析[J].实用医技杂志,2013,20(09):952-953. Chen J H,Liu B Y,Jia B X,et al.Diagnosis and analysis the reason of missed diagnosis and misdiagnosis of parathyroid adenoma by ultrasound[J].Journal of Practical Medical Techniques,2013,20(09):952-953.(in Chinese).
[14] 丁世飞,齐丙娟,谭红艳.支持向量机理论与算法研究综述[J].电子科技大学学报,2011,40(01):2-10. Ding S F,Qi B J,Tan H Y.An overview on theory and algorithm of support vector machines[J].Journal of University of Electronic Science and Technology of China,2011,40(01):2-10.(in Chinese).
[15] 杨金鸿,邓廷权.一种基于单簇核PCM的SVDD离群点检测方法[J].电子学报,2017,45(4):813-819. Yang J H,Deng T Q.A one-cluster kernel PCM based SVDD method for outliner detection[J].Acta Electronica Sinica,2017,45(4):813-819.(in Chinese).
[16] 陶新民,李晨曦,沈微,等.基于密度敏感最大软间隔SVDD不均衡数据分类算法[J].电子学报,2018,46(11):2725-2732. Tao X M,Li C X,Shen W,et al.The SVDD classifier for unbalanced data based on density-sensitive and maximum soft margin[J].Acta Electronica Sinica,2018,46(11):2725-2732.(in Chinese).
[17] 赵杰,祁永梅.一种新的甲状腺肿瘤超声图像特征提取算法[J].光电工程,2013,40(9):8-15. Zhao J,Qi Y M.A novel feature extraction algorithm of thyroid tumor ultrasound image[J].Opto-Electronic Engineering,2013,40(9):8-15.(in Chinese).
[18] Haralick R M,Shanmugam K,Dinstein I.Textural features for image classification[J].IEEE Transactions on Systems,Man,and Cybernetics,1973,SMC-3(6):610-621.
[19] 毛林,赵利强,于明安,等.基于图像局部熵的混合水平集模型甲状旁腺分割[J].光学学报,2019,39(12):256-264. Mao L,Zhao L Q,Yu M A,et al.Hybrid level set model for parathyroid gland segmentation based on local entropy of images[J].Acta Optica Sinica,2019,39(12):256-264.(in Chinese).
[20] Tolle C R,Mcjunkin T R,Rohrbaugh D T,et al.Lacunarity definition for ramified data sets based on optimal cover[J].Physica D:Nonlinear Phenomena,2003,179(3):129-152.
[21] 季桂树,禹智夫.分维和孔隙度在肝癌超声纹理识别中协同作用[J].计算机工程与应用,2013,49(02):211-218. Ji G S,Yu Z F.Synergy between fractal dimension and lacunarity discriminating liver cancer form normal liver from ultrasonic images[J].Computer Engineering and Applications,2013,49(02):211-218.(in Chinese).
[22] 康红霞,黄树彩,凌强,等.基于导弹尾焰特征谱的SVDD检测方法[J].红外技术,2015,37(08):696-700. Kang H X,Huang S C,Ling Q,et al.A detection method based on spectrum characteristics of missile plume using SVDD algorithm[J].Infrared Technology,2015,37(08):696-700.(in Chinese).
[23] Brereton R G.One-class classifiers[J].Journal of Chemometrics,2011,25(5):225-246.

融合先验知识特征的超声图像甲状旁腺结节识别

Recognition of Parathyroid Nodule by Fusing Prior Knowledge Features in Ultrasound Image

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

[1]	李春芳, 杨鑫, 张旭明, 丁明跃. 基于MAP的超声图像分解去噪算法研究[J]. 电子学报, 2014, 42(7): 1291-1298.
[2]	朱孝开;杨德贵. 基于推广能力测度的多类SVDD模式识别方法[J]. 电子学报, 2009, 37(3): 464-469.
[3]	付树军, 阮秋琦, 李玉, 王文洽. 基于各向异性扩散方程的超声图像去噪与边缘增强[J]. 电子学报, 2005, 33(7): 1191-1195.