北京生物医学工程

基于 Hurst 指数的脑胶质瘤分级方法

Grading method of gliomas based on Hurst index

作者：杨宇轩陶玲钱志余

单位：南京航空航天大学自动化学院(南京 211106)

关键词：脑胶质瘤；肿瘤分级；复杂度；赫斯特指数

分类号：R318.04

出版年·卷·期（页码）：2019·38·3（271-276）

摘要：

目的基于复杂度分析对胶质瘤患者的磁共振大脑静息态数据进行描述,寻找基于复杂度分析的肿瘤分级的客观指标? 方法基于复杂度赫斯特( Hurst)指数分析方法对被试者大脑肿瘤 fMRI影像的功能信息进行提取和分析,并对肿瘤进行分级研究? 首先基于 MRIcro 软件对患者肿瘤区域?对侧正常区域以及正常对照组的肿瘤对应区域进行提取;接着对提取出来的区域进行 Hurst 指数计算;然后对肿瘤区域及其对侧正常区域的 Hurst 指数值进行统计分析,对肿瘤区域及对照组同区域的 Hurst 指数值进行统计分析;最后将 29 例肿瘤患者样本按照病理等级进行分组,其中一级肿瘤患者 10 例,二级肿瘤患者 7 例,三?四级肿瘤患者各 6 例,对不同组别的 Hurst 指数进行双样本统计分析? 结果被试肿瘤区域的 Hurst 指数值与肿瘤等级成正比关系,肿瘤等级越高 Hurst 指数值越大? 统计分析表明不同级别肿瘤区域的 Hurst 指数差异具有统计学意义? 低级别肿瘤 Hurst 指数范围为 0.6381~0.6737,高级别肿瘤 Hurst 指数范围为 0.751 4 ~ 0.819 4? 结论 Hurst 指数分析方法可以区分低级别和高级别胶质瘤,可以为更细致的胶质瘤的等级划分提供帮助?

Objective To describe the MRI brain rest data of glioma patients based on complexity analysis,and to find objective indicators of tumor grading based on complexity analysis Methods Based on the complexity of the Hurst index analysis method,the functional information of fMRI images of brain tumors was extracted and analyzed,and the tumors were graded. Firstly,based on the MRIcro software,the correspondingregions of the tumor in the patient’ s tumor region,contralateral normal region,and normal control group were extracted;then the Hurst index was calculated for the extracted region;then the Hurst index value of the tumor region and its contralateral normal region was performed. Statistical analysis was performed on the Hurst index values in the same region of the tumor region and the control group. Finally,29 tumor patients were grouped according to pathological grade,including 10 primary tumor patients,7 secondary tumor patients,6 third? and fourth?grade tumor patients, and two?sample statistical analysis was performed on the Hurst index of different groups. Results The Hurst index value of the tumor region was proportional to the tumor grade. The higher the tumor grade was, the higher the Hurst index value was. The statistical analysis showed that the Hurst index of the tumor areas at different levels was significantly different. The Hurst index ranged from 0.6381 to 0.6737 in low?grade tumors,and from 0.7514 to 0.8194 in high?grade tumors. Conclusions The Hurst index analysis method can distinguish between low?grade and high?grade gliomas,and can provide help for more detailed classification of gliomas.

参考文献：

[ 1 ] 赵明,付旷,郭丽丽,等. 定量动态增强 MRI 在脑高低级别胶质瘤术前病理分级中的应用研究 [ J]. 中国实验诊断学,
2016,20(1):42-44.

Zhao M,Fu K,Guo LL,et al.Evaluation of T1?DCE MRI for the preoperative pathological evaluation of Tumor grade in brain
glioma [ J]. Chinese Journal of Laboratory Diagnosis, 2016, 20(1):42-44.
[ 2 ] Louis DN, Ohgaki H, Wiestler OD, et al. The 2007 WHO classification of tumours of the central nervous system[ J]. Acta Neuropathologica,2007,114 (2): 97-109.
[ 3 ] 范兵,杜华睿,王霄英,等. MRI 动态增强扫描定量参数对脑胶质瘤分级诊断价值的研究[ J]. 放射学实践,2014,29(8):
893-895.

Fan B,Du HR,Wang XY,et al. Dynamic enhanced MR scanning with volume transfer coefficient ( Ktrans ) in the evaluation of
histopathologic grading of cere?bral glioma [ J ]. Radiologic Practice,2014,29(8): 893-895.
[ 4 ] 张楠,杨本强. 脑胶质瘤分级诊断的磁共振研究新进展[ J].磁共振成像,2017,8(1):67-71.
Zhang N,Yang BQ. New approaches of magnetic resonance on the evaluation of brain glioma grading [ J ]. Chinese Journal of Magnetic Resonance Imaging,2017,8(1):67-71.
[ 5 ] Bertolaccini M, Bussolati C, Padovini G. A nonlinear filtering technique for the identification of biological signals [ J]. IEEE transactions on bio?medical engineering,1978,25(2):159-165.
[ 6 ] Smith RX, Yan L, Wang DJJ. Multiple time scale complexity analysis of resting state FMRI [ J]. Brain Imaging & Behavior,
2014,8(2):284-292.
[ 7 ] Wang Z,Li Y,Childress AR,et al. Brain entropy mapping using fMRI[J]. Plos One,2014,9(3): e89948.
[ 8 ] 张雅檬,陶玲,钱志余,等. 脑肿瘤患者大脑复杂度的 fMRI 研究[J].北京生物医学工程,2015,34(5):441-446.
Zhang YM,Tao L,Qian ZY,et al. A resting state fMRI study of brain complexity in cerebral tumor patients [ J ]. Beijing
Biomedical Engineering,2015,34(5):441-446.
[ 9 ] 白玲,景斌,叶德荣.大脑静息态 fMRI 信号复杂度的年龄效应研究[J]. 北京生物医学工程,2015,34(1):18-23.
Bai L, Jing B, Ye DR. Age effects on the complexity of brain endogenous oscillations in fMRI [ J ]. Beijing Biomedical Engineering,2015,34(1):18-23.
[10] Bernan J. Statistics for long?memory processes [M]. Boca Raton: Chapman & Hall / CRC,1994.
[11] Staff TPO. Correction: nonlinear complexity analysis of brain fMRI signals in Schizophrenia [ J ]. Plos One, 2014, 9 ( 8 ):
e95146.
[12] Maxim V, Sendur L, Fadili J, et al. Fractional gaussian noise, functional MRI and Alzheimer’s disease[J]. Neuroimage,2005, 25(1):141-158.
[13] Lai MC, Lombardo MV, Chakrabarti B, et al. A shift to randomness of brain oscillations in people with autism [ J ]. Biological Psychiatry,2010,68(12):1092-1099.
[14] Hahn T,Dresler T,Ehlis AC,et al. Randomness of resting?state brain oscillations encodes Gray ’ s personality trait [ J ]. Neuroimage,2012,59(2):1842-1845.
[15] Goldberger AL,Amaral LA,Hausdorff JM,et al. Fractal dynamics in physiology: alterations with disease and aging [ J ]

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录