北京生物医学工程

基于静息态fMRI的动态功能连接特征的智商预测

Prediction of intelligence quotient based on the characteristics of dynamic functional connectivity in resting state fMRI

作者：朱鸿睿

单位：北京交通大学计算机与信息技术学院（北京100044）

关键词：动态功能连接；智商；静息态fMRI ；回归

分类号：

出版年·卷·期（页码）：2020·39·1（62-68）

摘要：

目的近年来越来越多的研究表明大脑不同脑区间的功能连接的动态波动具有生理意义，但关于智商(intelligence quotient，IQ)的相关研究较少。本文基于动态功能连接（dynamic functional connections, DFC）提取动态特征对智商进行评估，为智商预测探索新的特征参数和预测模型。方法基于97个儿童静息态功能磁共振图像（resting state functional magnetic resonance image,RS-fMRI），采用滑动窗相关计算方法构建DFC。基于DFC提取相应时域、频域特征，通过弹性网（elastic-net，E-Net）和最小角回归（least angle regression，LAR）算法建立智商回归模型进行个体智商预测，并通过置换检验验证其显著性。结果基于动态功能连接的特定频段（0.075-0.1Hz）频域特征和波动均值(Mean)特征，可以实现对智商的基本预测，且频域特征的表现优于时域特征。另外，基于LAR算法构建的预测模型的表现优于E-Net算法。结论个体脑功能连接随时间的动态波动足以预测个体智商，且特定频段的频域特征和LAR算法能够提高预测准确率，这可为个体智商评估研究和动态功能连接的应用提供新的思路。

Objective In recent years, studies have shown that the dynamic fluctuations of functional connectivity can reflect physiological information, but few studies about intelligence quotient (IQ). This study evaluated IQ based on features extracted from dynamic functional connectivity (DFC), and new feature parameters and prediction models are explored for IQ prediction. Methods In the study resting state functional magnetic resonance imaging (RS-fMRI) data of 97 children subjects were selected, and the sliding window technique was used to constructed DFC. Based on DFC, the time domain and frequency domain characteristics were extracted, IQ regression models were established with the use of elastic net (E-Net) and least angle regression (LAR) algorithm, and its significance was verified by Permutation test. Results Individuals’ IQ could be predicted by frequency domain of 0.075-0.01Hz and the mean strength of dynamic fluctuation, the performance of frequency domain was better than that of time domain characteristics. Besides, the performance of the model based on the LAR algorithm was better than that of the E-Net algorithm. Conclusions IQ can be predicted by DFC characteristics, the frequency domain characteristics of specific frequency band and LAR algorithm can improve the prediction accuracy. This result provides new ideas for further research about IQ prediction and dynamic functional connectivity.

参考文献：

[1] Song M, Zhou Y, Li J, et al. Brain spontaneous functional connectivity and intelligence[J]. Neuroimage, 2008, 41(3):1168-1176.

[2] Langer N, Gianotti,LRR.,Knoch,D., et al.Functional brain network efficiency predicts intelligence[J].Human brain mapping,2012,33(6):1393-1406.

[3] Wang LY , Wee CY , Suk H , et al. MRI-based intelligence quotient (IQ) estimation with sparse learning[J]. Plos One, 2015, 10(3):e0117295-.

[4] Allen EA , Damaraju E , Plis SM , et al. Tracking whole-brain connectivity dynamics in the resting state[J]. Cerebral Cortex, 2014, 24(3):663-676.

[5] Liu J , Liao X , Xia M , et al. Chronnectome fingerprinting: Identifying individuals and predicting higher cognitive functions using dynamic brain connectivity patterns[J]. Human Brain Mapping, 2018, 39(2):902.

[6] Tian L , Li Q , Wang C , et al. Changes in dynamic functional connections with aging[J]. NeuroImage, 2018, 172:31-39.

[7] Hutchison RM , Womelsdorf T , Allen EA , et al. Dynamic functional connectivity: promise, issues, and interpretations[J]. NeuroImage, 2013, 80(1):360-378.

[8] Suk HI , Wee CY , Lee SW , et al. State-space model with deep learning for functional dynamics estimation in resting-state fMRI[J]. Neuroimage, 2016, 129:292-307.

[9] Consortium T A. The ADHD-200 consortium: a model to advance the translational potential of neuroimaging in clinical neuroscience[J]. Frontiers in Systems Neuroscience, 2012, 6:62.

[10] Dosenbach NU, Nardos B, Cohen A L, et al. Prediction of individual brain maturity using fMRI[J]. Science, 2010,329(5997):1358-1361

[11] Damaraju E , Allen EA , Belger A , et al. Dynamic functional connectivity analysis reveals transient states of dysconnectivity in schizophrenia[J]. NeuroImage: Clinical, 2014,5:298-308.

[12] Zalesky A, Breakspear M. Towards a statistical test for functional connectivity dynamics[J]. Neuroimage, 2015,114:466-470.

[13] Yaesoubi M , Allen EA , Miller RL , et al. Dynamic coherence analysis of resting fMRI data to jointly capture state-based phase, frequency, and time-domain information[J]. NeuroImage, 2015, 120:133-142.

[14] 赵欣, 张雄, 王伟伟, 等. 年龄相关的动态功能连接网络特征研究[J]. 生物医学工程学杂志, 2017(02):7-13.

Zhao X, Zhang X, Wang WW, et al. Research on the characteristics of the dynamic functional connectivity network related to aging[J]. Beijing Biomedical Engineering, 2017(02):7-13

[15] Zou H. The adaptive lasso and its oracle properties[J]. Journal of Industrial and Management Optimization (JIMO), 2006, 101(476):1418-1429.

[16] Khundrakpam BS , Tohka J , Evans AC . Prediction of brain maturity based on cortical thickness at different spatial resolutions[J]. NeuroImage, 2015, 111:350-359.

[17] Bunea F , She Y , Ombao H , et al. Penalized least squares regression methods and applications to neuroimaging[J]. Neuroimage, 2011, 55(4):1519-1527.

[18] Wager TD , Atlas LY , Leotti LA , et al. Predicting individual differences in placebo analgesia: contributions of brain activity during anticipation and pain experience[J]. Journal of Neuroscience, 2011, 31(2):439-452.

[19] Efron B, Hastie T, Johnstone I, et al. Least angle regression[J]. Annals of Statistics, 2004, 32(2):407-451.

[20] 皇甫浩然,杨剑,杨阳.基于fMRI动态功能连接的抑郁症患者分类研究[J].计算机应用研究, 2017,34(3):678-682.

Huangfu HR, Yang J,Yang Y. Classifying patients with depression based on fMRI dynamic functional connectivity[J]. Application Research of Computers, 2017,34(3):678-682.

[21] 蒋伟雄, 曾令李, 秦键, 等. 利用动态功能连接对健康危险性行为特征的预测[J]. 电子科技大学学报, 2018,47(6):927-931.

Jiang WX, Zeng LL, Qin J, et al. Prediction of the health-risk behavior by using dynamic functional connectivity[J]. Journal of University of Electronic Science and Technology of China, 2018,47(6):927-931.

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