Estimation and classification of brain functional networks based on temporal correlation information fusion

doi:10.11772/j.issn.1001-9081.2024050684

Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (5): 1403-1409.DOI: 10.11772/j.issn.1001-9081.2024050684

• China Conference on Data Mining 2024 (CCDM 2024) • Previous Articles

Estimation and classification of brain functional networks based on temporal correlation information fusion

Jun YANG¹, Mengxue PANG¹, Lishan QIAO¹^,²()

^1.School of Mathematical Science，Liaocheng University，Liaocheng Shandong 252000，China
^2.School of Computer Science and Technology，Shandong Jianzhu University，Jinan Shandong 250101，China

Received:2024-05-27 Revised:2024-06-22 Accepted:2024-06-28 Online:2024-07-25 Published:2025-05-10
Contact: Lishan QIAO
About author:YANG Jun， born in 2000， M. S. candidate. Her research interests include intelligent medicine， brain functional data analysis， machine learning.
PANG Mengxue， born in 1999， M. S. candidate. Her research interests include intelligent medicine， brain functional data analysis， machine learning.
QIAO Lishan， born in 1979. Ph. D.， professor. His research interests include artificial intelligence， pattern recognition， intelligent medicine， brain functional data analysis， machine learning.
Supported by:
National Natural Science Foundation of China — General Project(61976110);National Natural Science Foundation of China — Key Project(11931008)

融合时序相关信息的脑功能网络估计与分类

杨俊¹, 庞梦雪¹, 乔立山¹^,²()

^1.聊城大学数学科学学院，山东聊城 252000
^2.山东建筑大学计算机科学与技术学院，济南 250101

通讯作者: 乔立山
作者简介:杨俊（2000—），女，山东新泰人，硕士研究生，主要研究方向：智能医学、脑功能数据分析、机器学习
庞梦雪（1999—），女，山东惠民人，硕士研究生，主要研究方向：智能医学、脑功能数据分析、机器学习
乔立山（1979—），男，山东邹平人，教授，博士，CCF会员，主要研究方向：人工智能、模式识别、智能医学、脑功能数据分析、机器学习。
基金资助:
国家自然科学基金面上项目(61976110);国家自然科学基金重点项目(11931008)

Abstract

Abstract:

Brain functional networks play a crucial role in the early diagnosis of neurological or encephaloid diseases， and the estimation of a high-quality brain functional network is one of the most critical challenges. Although numerous brain functional network estimation methods have been proposed， most of them only focus on the correlations among brain regions while ignoring potential dependencies among time points. Recent studies have found that encoding dependencies among time points can effectively improve the discriminative properties of brain functional networks； however， this method only relies on the dependencies between adjacent time points and fails to effectively utilize information from non-adjacent time points， thereby inadequately capturing the temporal characteristics of the brain functional network. To address this limitation， a new brain functional network estimation method was proposed， which introduced a similarity matrix to encode the dependencies among non-adjacent time points， aiming to improve the quality of the estimation. Additionally， an alternating optimization learning algorithm was designed to solve the model quickly. To evaluate the effectiveness of the proposed method， experiments were conducted on three public datasets — ADNI （Alzheimer's Disease Neuroimaging Initiative）， ABIDE （Autism Brain Imaging Data Exchange）， and REST-MDD （REST-meta-MDD Consortium） — for mild cognitive impairment， autism and depression， respectively. Experimental results demonstrate that the brain functional network estimated by the proposed method achieves superior classification performance.

Key words: brain functional network, similarity matrix, latent variable, resting-state functional Magnetic Resonance Imaging (rs-fMRI), temporal information

摘要：

脑功能网络在神经或精神类脑疾病的早期诊断中发挥着重要作用，而估计一个高质量的脑功能网络是其中最关键的问题之一。尽管目前已有众多脑功能网络估计方法，但多数仅考虑了脑区间的相关性，忽视了时间点间可能存在的依赖关系。最近的研究发现，引入潜变量编码时间点间的依赖性可以有效提高脑功能网络的判别性；但该方法仅基于相邻时间点的依赖关系，并未有效利用不相邻时间点的信息，无法全面反映脑功能网络的时序特性。因此，提出一种新的脑功能网络估计方法，通过引入相似性矩阵编码不相邻时间点间的依赖关系，旨在提高脑功能网络估计的质量；并设计了交替优化学习算法快速求解该方法的模型。为了评估所提方法的有效性，在3个公开数据集ADNI（Alzheimer's Disease Neuroimaging Initiative）、ABIDE（Autism Brain Imaging Data Exchange）和REST-MDD（REST-meta-MDD Consortium）上分别进行了轻度认知障碍、孤独症与抑郁症的识别实验，实验结果表明，基于所提方法估计的脑功能网络能够获得更优的分类性能。

关键词: 脑功能网络, 相似性矩阵, 隐变量, 静息态功能磁共振成像, 时序信息

CLC Number:

TP391

Jun YANG, Mengxue PANG, Lishan QIAO. Estimation and classification of brain functional networks based on temporal correlation information fusion[J]. Journal of Computer Applications, 2025, 45(5): 1403-1409.

杨俊, 庞梦雪, 乔立山. 融合时序相关信息的脑功能网络估计与分类[J]. 《计算机应用》唯一官方网站, 2025, 45(5): 1403-1409.

Figures/Tables 11

Fig. 1 Differences between I-SRiLT and SRiLT in encoding temporal dependencies

Tab. 1 Demographic and clinical information of subjects in ADNI dataset

类别	性别（男/女）	年龄	MMSE
MCI（ N=68）	39/29	76.50±13.50	26.77±1.23
NC（ N=69）	17/52	71.50±14.50	28.85±1.15

Fig. 2 Flow of feature selection and classification

Tab. 2 Definition of performance evaluation indicators

性能评价指标	定义
准确度	$(T P + T N) / (T P + F P + T N + F N)$
灵敏度	$T P / (T P + F N)$
特异性	$T N / (T N + F P)$
平衡率	$(S E N + S P E) / 2$
阳性查准率	$T P / (T P + F P)$
阴性查准率	$T N / (T N + F N)$
F1分数	$2 T P / (T P + T P + F N + F P)$

Tab. 2 Definition of performance evaluation indicators

性能评价指标	定义
准确度	$(T P + T N) / (T P + F P + T N + F N)$
灵敏度	$T P / (T P + F N)$
特异性	$T N / (T N + F P)$
平衡率	$(S E N + S P E) / 2$
阳性查准率	$T P / (T P + F P)$
阴性查准率	$T N / (T N + F N)$
F1分数	$2 T P / (T P + T P + F N + F P)$

Tab. 1 Classification performance comparison of different methods on ADNI dataset

方法	准确度	灵敏度	特异性	平衡率	阳性查准率	阴性查准率	F1分数	AUC
PC	0.671 5	0.720 6	0.623 2	0.671 9	0.653 3	0.693 5	0.685 3	0.730 2
SR	0.715 3	0.720 6	0.710 1	0.715 4	0.710 1	0.720 6	0.715 3	0.783 9
SR+SS	0.737 2	0.764 7	0.710 1	0.737 4	0.722 2	0.753 8	0.742 9	0.773 2
SR+W	0.744 5	0.750 0	0.739 1	0.744 6	0.739 1	0.750 0	0.739 1	0.796 9
SRiLT	0.839 4	0.852 9	0.826 1	0.839 5	0.828 6	0.850 7	0.840 6	0.929 0
I-SRiLT	0.854 0	0.838 2	0.869 6	0.853 9	0.863 6	0.845 1	0.850 7	0.927 3

Fig. 3 ROC curves of different methods on ADNI dataset

Fig. 4 Classification accuracies of different methods under different parameter values

Fig. 5 Selected top 22 most discriminative functional connections using t-test

Tab. 4 Demographic information and fMRI parameters of ABIDE and REST-MDD datasets

数据信息	患者（男/女）	正常（男/女）	年龄	TR/ms	TE/ms	Thickness/mm	TP	Flip Angle	Slice Number
ABIDE （NYU）	79（68/11）	105（79/26）	14.52±6.97	2 000	15	4.0	180	90°	33
REST-MDD （Site20）	282（99/183）	251（87/164）	38.74±13.74	2 000	30	3.0	242	90°	32

Tab. 5 Classification performance of different methods on ABIDE dataset

方法	准确度	灵敏度	特异性	平衡率	阳性查准率	阴性查准率	F1分数	AUC
PC	0.608 8	0.582 5	0.629 4	0.606 0	0.570 7	0.655 4	0.548 3	0.685 2
SR	0.619 9	0.436 6	0.782 4	0.609 5	0.589 6	0.670 1	0.463 2	0.630 9
SR+SS	0.668 5	0.658 2	0.676 2	0.667 2	0.604 7	0.724 5	0.630 6	0.721 9
SR+W	0.679 8	0.636 9	0.717 4	0.677 1	0.636 4	0.717 6	0.626 6	0.737 9
SRiLT	0.684 8	0.620 3	0.733 3	0.676 8	0.636 4	0.719 6	0.628 2	0.731 2
I-SRiLT	0.706 4	0.614 5	0.759 7	0.687 1	0.661 4	0.720 5	0.626 1	0.756 2

Tab. 6 Classification performance of different methods on REST-MDD dataset

方法	准确度	灵敏度	特异性	平衡率	阳性查准率	阴性查准率	F1分数	AUC
PC	0.568 5	0.588 9	0.541 1	0.565 0	0.592 3	0.538 0	0.588 3	0.608 7
SR	0.542 3	0.607 7	0.465 5	0.536 6	0.558 1	0.518 8	0.579 6	0.554 4
SR+SS	0.551 6	0.602 4	0.485 7	0.544 1	0.569 3	0.523 8	0.583 1	0.562 0
SR+W	0.570 3	0.600 3	0.539 6	0.569 9	0.590 7	0.554 7	0.590 6	0.581 2
SRiLT	0.546 3	0.613 1	0.477 6	0.545 4	0.568 2	0.525 7	0.584 8	0.559 9
I-SRiLT	0.605 6	0.615 4	0.569 3	0.592 3	0.620 2	0.566 2	0.613 6	0.626 5

References 25

1	McCARTY P J， PINES A R， SUSSMAN B L， et al. Resting state functional magnetic resonance imaging elucidates neurotransmitter deficiency in autism spectrum disorder［J］. Journal of Personalized Medicine， 2021， 11（ 10）： No. 969.
2	ZHU X， YUAN F， ZHOU G， et al. Cross-network interaction for diagnosis of major depressive disorder based on resting state functional connectivity［J］. Brain Imaging and Behavior， 2021， 15（ 3）： 1279- 1289.
3	余仁萍，余海飞，万红. 基于静息态功能磁共振成像的精神分裂症脑网络特征分类研究［J］. 生物医学工程学杂志， 2020， 37（ 4）： 661- 669.
	YU R P， YU H F， WAN H. Research on brain network for schizophrenia classification based on resting-state functional magnetic resonance imaging［J］. Journal of Biomedical Engineering， 2020， 37（ 4）： 661- 669.
4	LI G， LIU Y， ZHENG Y， et al. Large-scale dynamic causal modeling of major depressive disorder based on resting-state functional magnetic resonance imaging［J］. Human Brain Mapping， 2020， 41（ 4）： 865- 881.
5	JENKINSON M， CHAPPELL M. Introduction to neuroimaging analysis［M］. Oxford： Oxford University Press， 2018.
6	QIU Y-H， HUANG Z-H， GAO Y-Y， et al. Alterations in intrinsic functional networks in Parkinson's disease patients with depression： a resting-state functional magnetic resonance imaging study［J］. CNS Neuroscience and Therapeutics， 2021， 27（ 3）： 289- 298.
7	ZHAO F， ZHANG H， REKIK I， et al. Diagnosis of autism spectrum disorders using multi-level high-order functional networks derived from resting-state functional MRI［J］. Frontiers in Human Neuroscience， 2018， 12： No. 184.
8	RONICKO J F A， THOMAS J， THANGAVEL P， et al. Diagnostic classification of autism using resting-state fMRI data improves with full correlation functional brain connectivity compared to partial correlation［J］. Journal of Neuroscience Methods， 2020， 345： No. 108884.
9	MAHADEVAN A S， TOOLEY U A， BERTOLERO M A， et al. Evaluating the sensitivity of functional connectivity measures to motion artifact in resting-state fMRI data［J］. NeuroImage， 2021， 241： No. 118408.
10	LI W， ZHANG L， QIAO L， et al. Toward a better estimation of functional brain network for mild cognitive impairment identification： a transfer learning view［J］. IEEE Journal of Biomedical and Health Informatics， 2020， 24（ 4）： 1160- 1168.
11	QIAO L， ZHANG L， CHEN S， et al. Data-driven graph construction and graph learning： a review［J］. Neurocomputing， 2018， 312： 336- 351.
12	BIJSTERBOSCH J， SMITH S， BECKMANN C F. Introduction to resting state fMRI functional connectivity［M］. Oxford： Oxford University Press， 2017.
13	LI W， WANG Z， ZHANG L， et al. Remodeling Pearson's correlation for functional brain network estimation and autism spectrum disorder identification［J］. Frontiers in Neuroinformatics， 2017， 11： No. 55.
14	QIAO L， ZHANG H， KIM M， et al. Estimating functional brain networks by incorporating a modularity prior［J］. NeuroImage， 2016， 141： 399- 407.
15	WANG J， ZHANG L， WANG Q， et al. Multi-class ASD classification based on functional connectivity and functional correlation tensor via multi-source domain adaptation and multi-view sparse representation［J］. IEEE Transactions on Medical Imaging， 2020， 39（ 10）： 3137- 3147.
16	XUE Y， ZHANG Y， ZHANG L， et al. Learning brain functional networks with latent temporal dependency for MCI identification［J］. IEEE Transactions on Biomedical Engineering， 2022， 69（ 2）： 590- 601.
17	LI W， QIAO L， ZHANG L， et al. Functional brain network estimation with time series self-scrubbing［J］. IEEE Journal of Biomedical and Health Informatics， 2019， 23（ 6）： 2494- 2504.
18	SU H， ZHANG L， QIAO L， et al. Estimating high-order brain functional networks by correlation-preserving embedding［J］. Medical and Biological Engineering and Computing， 2022， 60（ 10）： 2813- 2823.
19	COMBETTES P L， PESQUET J C. Proximal splitting methods in signal processing［M］// Fixed-point algorithms for inverse problems in science and engineering， SOIA 49. New York： Springer， 2011： 185- 212.
20	TZOURIO-MAZOYER N， LAMDEAU B， PAPATHANASSIOU D， et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain［J］. NeuroImage， 2002， 15（ 1）： 273- 289.
21	CHEN H， ZHANG Y， ZHANG L， et al. Estimating brain functional networks based on adaptively-weighted fMRI signals for MCI identification［J］. Frontiers in Aging Neuroscience， 2020， 12： No. 595322.
22	SCHULTZ S A， BOOTS E A， DARST B F， et al. Cardiorespiratory fitness alters the influence of a polygenic risk score on biomarkers of AD［J］. Neurology， 2017， 88（ 17）： 1650- 1658.
23	PAN P， ZHU L， YU T， et al. Aberrant spontaneous low-frequency brain activity in amnestic mild cognitive impairment： a meta-analysis of resting-state fMRI studies［J］. Ageing Research Reviews， 2017， 35： 12- 21.
24	GAO Y， NIE K， HUANG B， et al. Changes of brain structure in Parkinson's disease patients with mild cognitive impairment analyzed via VBM technology［J］. Neuroscience Letters， 2017， 658： 121- 132.
25	BAILLY M， DESTRIEUX C， HOMMET C， et al. Precuneus and cingulate cortex atrophy and hypometabolism in patients with Alzheimer's disease and mild cognitive impairment： MRI and ¹⁸F‑FDG PET quantitative analysis using FreeSurfer［J］. BioMed Research International， 2015， 2015： No. 583931.

[1]	Jingxin LIU, Wenjing HUANG, Liangsheng XU, Chong HUANG, Jiansheng WU. Unsupervised feature selection model with dictionary learning and sample correlation preservation [J]. Journal of Computer Applications, 2024, 44(12): 3766-3775.
[2]	Junhong ZHU, Junyu LAI, Lianqiang GAN, Zhiyong CHEN, Huashuo LIU, Guoyao XU. Video prediction model combining involution and convolution operators [J]. Journal of Computer Applications, 2024, 44(1): 113-122.
[3]	Tian HE, Zongxin SHEN, Qianqian HUANG, Yanyong HUANG. Adaptive learning-based multi-view unsupervised feature selection method [J]. Journal of Computer Applications, 2023, 43(9): 2657-2664.
[4]	Jianwen GAN, Yan CHEN, Peng ZHOU, Liang DU. Clustering ensemble algorithm with high-order consistency learning [J]. Journal of Computer Applications, 2023, 43(9): 2665-2672.
[5]	Ranyan NI, Yi ZHANG. Action recognition method based on video spatio-temporal features [J]. Journal of Computer Applications, 2023, 43(2): 521-528.
[6]	Zhifeng MA, Junyang YU, Longge WANG. Diversity represented deep subspace clustering algorithm [J]. Journal of Computer Applications, 2023, 43(2): 407-412.
[7]	Yu YANG, Weiwei DUAN. Spectral clustering based dynamic community discovery algorithm in social network [J]. Journal of Computer Applications, 2023, 43(10): 3129-3135.
[8]	ZHANG Cheng, WAN Yuan, QIANG Haopeng. Deep unsupervised discrete cross-modal hashing based on knowledge distillation [J]. Journal of Computer Applications, 2021, 41(9): 2523-2531.
[9]	LIU Ziyan, ZHU Mingcheng, YUAN Lei, MA Shanshan, CHEN Lingzhouting. Video person re-identification based on non-local attention and multi-feature fusion [J]. Journal of Computer Applications, 2021, 41(2): 530-536.
[10]	KONG Lingxu, WU Haifeng, ZENG Yu, LU Xiaoling. Diagnosis of mild cognitive impairment using deep learning and brain functional connectivities with different frequency dimensions [J]. Journal of Computer Applications, 2021, 41(2): 590-597.
[11]	LI Hanlu, XIE Qing, TANG Lingli, LIU Yongjian. POI recommendation algorithm combining spatiotemporal information and POI importance [J]. Journal of Computer Applications, 2020, 40(9): 2600-2605.
[12]	LYU Yali, MIAO Junzhong, HU Weixin. Semi-supervised learning algorithm of graph based on label metric learning [J]. Journal of Computer Applications, 2020, 40(12): 3430-3436.
[13]	SONG Yan, YIN Jun. Multi-view spectral clustering algorithm based on shared nearest neighbor [J]. Journal of Computer Applications, 2020, 40(11): 3211-3216.
[14]	YANG Yanlin, YE Zhonglin, ZHAO Haixing, MENG Lei. Link prediction algorithm based on high-order proximity approximation [J]. Journal of Computer Applications, 2019, 39(8): 2366-2373.
[15]	DING Chao, ZHAO Hai, SI Shuaizong, ZHU Jian. Evolution model of normal aging human brain functional network [J]. Journal of Computer Applications, 2019, 39(4): 963-971.

Estimation and classification of brain functional networks based on temporal correlation information fusion

融合时序相关信息的脑功能网络估计与分类

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 11

References 25

Related Articles 15

Recommended Articles

Metrics