北京生物医学工程

基于功能连接的大脑年龄预测的影响因素

Influential factors of brain age prediction based on functional connectivity

作者：周震王洪景斌

单位：首都医科大学生物医学工程学院（北京 100069），通讯作者：周震。E-mail:zhouzhenbme@126.com

关键词：年龄预测; 功能连接; 大脑模板; 全局信号回归; 支持向量回归

分类号：R318. 04

出版年·卷·期（页码）：2021·40·4（400-405）

摘要：

目的探索基于静息态功能连接进行大脑生理年龄预测的可行性及相关影响因素。方法选取来自阿尔茨海默病神经影像学计划( Alzheimer disease neuroimaging initiative , ADNI)数据库的41例满足条件的健康受试者。首先对静息态功能磁共振成像( resting- state functional MRI, rs-fMRI)数据进行预处理并提取功能连接特征，利用基于Bootstrap的特征筛选方法进行特征降维;然后使用支持向量回归建立正常人大脑年龄的预测模型,最后用留一法进行交叉验证,并比较不同大脑模板、全脑信号回归及性别因素对年龄预测的影响。结果基于AAL-90、AAL-1024、Shen-268、Fan-246脑图谱得到的预测值与真实年龄之间的相关系数r分别为0.23、0.29 ,0.17.0. 38。使用全局信号回归,基于Fan-246脑图谱得到年龄预测模型的相关系数r显著提升为0.66。利用性别分组建模,基于Fan-246脑图谱预测模型的相关系数r提升为0.46。结论根据静息态功能磁共振的功能连接特征可以较好地估计健康大脑的生理年龄,且大脑模板、全局信号回归对年龄估计模型的性能有较大影响。本研究可加深对大脑老化过程的认识，对阿尔兹海默病的早期诊断和预防有着重要的指导价值。

Object To explore the feasibility and related influential factors in brain age prediction basedon resting· -state functional connectivity. Methods Forty-one eligible healthy subjects were enrolled from the opensource Alzheimer disease neuroimaging initiative ( ADNI) dataset in this study. After the preprocess of resting-state functional MRI( rs-fMRI) data and the extraction of functional connectivity, the feature selection methodbased on Bootstrap was used for feature reduction , which was then input into the support vector regression modelto construct the age estimation model of healthy brains. The leave-one out cross validation was applied for modelevaluation , and the influences of different brain atlases , global signal regression and gender on the estimationperformances were also compared. Results The correlational values between the estimated age and the real agewere 0.23 ,0.29 ,0.17,0.38 respectively for AAL-90, AAL- 1024 , Shen 268 and Fan- 246 atlases. When using information to construct the gender· -specific age estimation model, the performance of Fan- 246 atlas couldincrease to 0.46. Conclusions The functional connectivity features from resting: -state functional MRI can wellestimate the physiological age of healthy brains , and brain atlas and global signal regression are important factorsfor the age estimation model. This study can deepen the understanding of the aging process in the brain , andprovide valuable guidance to the early detection and prevention of Alzheimer disease.

参考文献：

[1] Cole JH,Franke K. Predicting age using neuroimaging : innovative brain ageing biomarkers[J]. Trends in Neuroscience, 2017, 40(12) :681-690.
[2] Daffner KR. Promoting uccessful cognitive aging:a comprehensive review[J]. Journal of Alzheimers' Disease , 2010, 19(4):1101-1122.
[3] Luders E, Cherbuin N , Gaser C. E5stimating brain age using highresolution pattermrecognition; Y ounger brains in long-termmeditation pratitioners [ J]. Neurolmage, 2016, 134(134):508-513.
[4] Lin I.,Jin C, Fu Z.,el al. Predieting healthy older adull' s brain age based on structural connectivity networks using artilicialneural networks [ J ]. Computer Methods and Programs inBiomedicine ,2016,20(12):8-17.
[5] HanCE, Peraza LR, 'aylor J,et al. Predicting age of human subjecls based on struetural connectivily from diffusion tensorimaging[ J]. Neurolmage ,2014,21(4):11-17.
[6] Liem F, Varoquaux G, Kynast J,et al. Predieting brain-age from multimodal imaging data eaptures cognitive impaimment [J] .Neurolmage. .2017, 148(21):179- 188.
[7] Dosenbach NU, Nardos B, Cohen AL, et al. Prediction of individual brain maturity using fMRI[J]. Science, 2010, 329(5997): 1358- 1361.
[8] Laneaster J, Loranz R, Leech R,et al. Bayesian optimization for neuroimaging pre-processing in brain age classification andprediction[ J ]. Frontiers in Aging Neuroscience, 2018, 10:10-28.
[9] Li H, Stterthwaite TD, Fan Y. Brain age prediction based on Resting- state functional connectivity patterns using convolutionalneural networks[ J ]. IEEE International Symposium on Biomedical lmaging, 2018, 102: 101-104.
[10] Wei LJ,Jing B, Li HY . Bootstrapping promoles the RSFCbehavior associations : An application of individual cognitive traitspredietion[ J]. Human Brain Mapping, 2020,41(9) :2302-2316.
[11] Liu TT, Nalci A, Falahpour M. 'The global signal in fMRl:nuisance or information [J]. Neurolmage, 2017, 150 (20):213-229.
[12] Power JD, Plitt M , Gotts SJ , et al. Ridding fMRI data of motion related influences: removal of signals with distinct spatial andphysical bases in multiecho d?ta[ J]. Proceedings of the NátionalAcademy of Sciences of the United States of America ,2018,115(9): E2105-E2114.
[13]Satterthwaite TD, Elliott MA, Gerraty RT, et al. An improvedframework for confound regression and filtering for control ofmotion artifact in the preprocessing of resting- -state functionalconnectivity data[J]. Neurolmage ,2013 ,64(20) :240-256
[14]Tzourio-Mazoyer N, Landeau B, Papathanassiou D, et al.Automated anatomical labeling of activations in SPM using amacroscopic anatomical parellation of the MNI MRI single-subject brain[ J]. Neurolmage .2002,15(1) :273-289.
[15] Finn ES, Shen X, Scheinost D, et al. Functional connectome fingerprinting: identifying individuals using patterns of brainconnectivity [J]. Nature Neumscience, 2015, 18 (11):1664-1671.
[16] Fan LZ,Li H,Zhuo JJ,et al. The human brainnetome atlas:a new brain atlas based on connectional architecture [ J ]. Cerebral .Cortex ,2016 ,26( 8) :3508-3526.
[17]Jiang RT, Calhoun VD, Fan LZ, et al. Gender differences inconnectome-based predictions of individualized inelligencequotient and sub-domain scores [ J]. Cerebral Cortex , 2020, 30(3) : 88-900.
[18] Andrús Király, Nikoletta Szabó, Eszter Tóth, et al. Male brain ages faster : the age and gender dependence of subcortical volumes[J]. Brain Imaging & Bchavior , 2016, 10(3) :901-910.

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