Sun JJ, Zhang L, Sun RH, Gao XZ, Fang CX, Zhou ZH. Identification of key brain networks and functional connectivities of successful aging: A surface-based resting-state functional magnetic resonance study. World J Psychiatry 2025; 15(3): 100456 [DOI: 10.5498/wjp.v15.i3.100456]
Corresponding Author of This Article
Zhen-He Zhou, MD, Professor, Department of Psychiatry, The Affiliated Mental Health Center of Jiangnan University, No. 156 Qianrong Road, Wuxi 214151, Jiangsu Province, China. zhouzh@njmu.edu.cn
Research Domain of This Article
Gerontology
Article-Type of This Article
Observational Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
World J Psychiatry. Mar 19, 2025; 15(3): 100456 Published online Mar 19, 2025. doi: 10.5498/wjp.v15.i3.100456
Identification of key brain networks and functional connectivities of successful aging: A surface-based resting-state functional magnetic resonance study
Jiao-Jiao Sun, Xue-Zheng Gao, Chun-Xia Fang, Zhen-He Zhou, Department of Psychiatry, The Affiliated Mental Health Center of Jiangnan University, Wuxi 214151, Jiangsu Province, China
Jiao-Jiao Sun, Ru-Hong Sun, Department of Psychiatry, Yangzhou Wutaishan Hospital of Jiangsu Province, Teaching Hospital of Yangzhou University, Yangzhou 225000, Jiangsu Province, China
Li Zhang, Department of Psychiatry, Huai’an Third People’s Hospital, Huai’an 223300, Jiangsu Province, China
Co-corresponding authors: Chun-Xia Fang and Zhen-He Zhou.
Author contributions: Sun JJ and Zhang L contributed to the software of the manuscript, they contributed equally to this article, they are the co-first authors of this manuscript; Sun RH and Gao XZ investigated and resourced the manuscript; Sun JJ, Zhang L, Sun RH, and Gao XZ wrote the original manuscript; Fang CX contributed to the conceptualization, methodology, data curation, writing, visualization, project management, and acquisition of funds; Sun JJ, Zhang L, and Fang CX formally analyzed the manuscript; Zhou ZH contributed to the manuscript with resources, data organization, and supervision; Fang CX and Zhou ZH they contributed equally to this article, they are the co-corresponding authors of this manuscript; and all authors contributed to the article and approved the submitted version.
Supported by the Wuxi Municipal Health Commission Major Project, No. Z202107.
Institutional review board statement: This study was approved by the Medical Ethics Committee of the Wuxi Mental Health Center, approval No. WXMHCIRB2017LL07; and all procedures performed in this study involving human participants were in accordance with the Declaration of Helsinki.
Informed consent statement: All participants enrolled into this study provided informed written consent prior to study enrollment.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
STROBE statement: The authors have read the STROBE Statement-checklist of items, and the manuscript was prepared and revised according to the STROBE Statement-checklist of items.
Data sharing statement: Data used in this study can be available from the corresponding author at zhouzh@njmu.edu.cn.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Zhen-He Zhou, MD, Professor, Department of Psychiatry, The Affiliated Mental Health Center of Jiangnan University, No. 156 Qianrong Road, Wuxi 214151, Jiangsu Province, China. zhouzh@njmu.edu.cn
Received: August 17, 2024 Revised: January 11, 2025 Accepted: January 22, 2025 Published online: March 19, 2025 Processing time: 192 Days and 14.7 Hours
Abstract
BACKGROUND
Successful aging (SA) refers to the ability to maintain high levels of physical, cognitive, psychological, and social engagement in old age, with high cognitive function being the key to achieving SA.
AIM
To explore the potential characteristics of the brain network and functional connectivity (FC) of SA.
METHODS
Twenty-six SA individuals and 47 usual aging individuals were recruited from community-dwelling elderly, which were taken the magnetic resonance imaging scan and the global cognitive function assessment by Mini Mental State Examination (MMSE). The resting state-functional magnetic resonance imaging data were preprocessed by DPABISurf, and the brain functional network was conducted by DPABINet. The support vector machine model was constructed with altered functional connectivities to evaluate the identification value of SA.
RESULTS
The results found that the 6 inter-network FCs of 5 brain networks were significantly altered and related to MMSE performance. The FC of the right orbital part of the middle frontal gyrus and right angular gyrus was mostly increased and positively related to MMSE score, and the FC of the right supramarginal gyrus and right temporal pole: Middle temporal gyrus was the only one decreased and negatively related to MMSE score. All 17 significantly altered FCs of SA were taken into the support vector machine model, and the area under the curve was 0.895.
CONCLUSION
The identification of key brain networks and FC of SA could help us better understand the brain mechanism and further explore neuroimaging biomarkers of SA.
Core Tip: This study investigates the brain network characteristics and functional connectivity (FC) associated with successful aging (SA) using resting-state functional magnetic resonance imaging. The results found that the 6 inter-network FC of 5 brain networks were significantly altered and related to Mini Mental State Examination performance, of which the default mode network, attention network, and language network were the most concentrated networks. The identification of key brain networks and FC of SA could help us better understand the brain mechanism and further explore neuroimaging biomarkers of SA.
Citation: Sun JJ, Zhang L, Sun RH, Gao XZ, Fang CX, Zhou ZH. Identification of key brain networks and functional connectivities of successful aging: A surface-based resting-state functional magnetic resonance study. World J Psychiatry 2025; 15(3): 100456
To date, most studies have attempted to understand the aging process from the perspective of diseases, such as neurodegenerative diseases[1,2]. However, approximately 8.5% (in Europe) to 25.4% (in Singapore) of elderly individuals around the world can still maintain high levels of cognitive and physical function in late life[3], which was first referred to as successful aging (SA)[4-6]. However, there is no consensus on the definition and criteria of SA, and many researchers have tried to expound its connotation successively[7,8] from the three-dimensional model[4] to the five-dimensional model[9] of physiology, society, psychology, etc. Regardless of the field or dimension, maintaining high levels of cognitive function, so-called successful cognitive aging[10], has been recognized as a crucial feature of SA[11].
The application of neuroimaging technology makes it a good attempt to explore the brain mechanism and objective biological indicators of SA. However, there are few structural and functional magnetic resonance imaging (MRI) studies on the neuroimaging of SA, and most of the imaging research has focused on the cognitive field, such as “supernormals”[12,13] and “superagers”[14,15]. Both are branches of “successful cognitive aging” research. On the basis of the Alzheimer's disease neuroimaging initiative (ADNI) program, several research groups in America characterizes “supernormals” as elderly individuals whose trend of changes in cognitive function within 5 years is better than that of their peers[12,13]. Meanwhile, several research groups in Europe have defined “superagers” as elderly individuals whose memory and executive functions within cognitive performance are comparable to those of individuals aged 18-32 years[14,15]. Studies have shown that superagers exhibit greater cortical thickness in areas like the anterior cingulate cortex (ACC), which may help resist age-related cognitive decline[16,17]. Research by Sun et al[14] also found that superagers have thicker regions in the default mode and salience networks, such as the anterior temporal cortex and hippocampus, correlating with better memory. Similarly, studies on supernormals have identified regions like the right fusiform gyrus and ACC with greater resistance to age-related neurodegeneration[12]. In our previous study, we used volume-based resting state-functional MRI (rs-fMRI) to analyze the brain functional features of SA and found that the brain regions with altered functional MRI (fMRI) characteristics amplitude of low frequency fluctuations (ALFF) (degree centrality; regional homogeneity (ReHo); voxel-mirrored homotopic connectivity) in the SA group were concentrated in the frontal (6 brain regions) and temporal (4 brain regions) lobes. ALFF in the right opercular part of the inferior frontal gyrus was significantly correlated with cognitive function and age, which might be used to distinguish SA[18].
Meanwhile, only 3 studies have explored the functional connectivity (FC) or brain network of supernomals and superagers instead of SA. Using fMRI data from the ADNI program, Lin et al[10] found markedly more significantly increased FC between the ACC and the right hippocampus, the middle cingulate cortex (MCC) and the left super temporal gyrus, and the posterior cingulate cortex and the right precuneus in the over-achievers, and weaker FC between the right middle frontal gyrus and the MCC , as well as between the right thalamus and the MCC. All these FCs were significantly related to memory and overall cognitive performance across all participants. Utilizing diffusion tensor imaging data from the ADNI, Chen et al[13] found a unique structural connectome containing connections among frontal, cingulate, parietal, temporal, and subcortical regions in the same hemisphere that remained stable over time in supernormals; in addition, the mean diffusivity in these individuals between the right isthmus cingulate cortex and right precuneus could predict Alzheimer’s disease pathology. Zhang et al[19] revealed that superagers exhibited greater connectivity within the default mode network (DMN) and the significance network compared to typical older adults, and their connectivity in these networks was similar to that of young adults. Notably, the greater the connectivity within each network, the better the older adults performed on the California Verbal Learning Test and recognition tasks. Therefore, structural and functional MRI studies of supernormals and superagers may have their own emphasis and have few similarities. While the definitions of supernormals and superagers are limited to memory and executive function of cognitive function, the brain imaging mechanism of SA may be more complex[13,14].
Brain network analysis, which uses mathematical methods to reveal the connectivity and organization patterns of the whole brain based on graphics or networks[20,21], offering a more comprehensive view of brain function than traditional methods that focus on isolated regions. In this study, we applied surface-based network analysis to explore the FC characteristics of the global brain network in SA. By adopting this method, we aimed to better understand how brain regions interact within a network, providing deeper insights into the complex neural mechanisms of SA. Additionally, we used the support vector machine (SVM) algorithm to develop a model based on differential FC, assessing its potential to distinguish SA from usual aging (UA). This approach not only allows for a more holistic understanding of brain connectivity but also offers a powerful tool for identifying specific brain network features unique to SA.
MATERIALS AND METHODS
Participants
From April to October 2018, this study enrolled 26 participants in the SA group and 47 participants in the UA group, both groups consisting of community-dwelling elderly individuals from Wuxi City, China. This dataset has been previously utilized by our group[18]. The detailed inclusion and exclusion criteria are shown in the Supplementary material. This study was approved by the Ethics Committee of the Wuxi Mental Health Center. All participants provided written informed consent prior to participation, and all procedures performed in this study involving human participants were in accordance with the Declaration of Helsinki.
Neuropsychological tests
The Mini Mental State Examination (MMSE) was used to assess the global cognitive function of all the subjects before MRI scanning was conducted. The MMSE is a widely used tool in clinical practice to quickly assess the cognitive status of elderly people and has high specificity and sensitivity when first tested[22]. The activity of daily living scale was used to assess the daily living capacity of all the subjects.
MRI scan
The detailed MRI scanning parameters were the same as those shown in a previous article and are also shown in the Supplementary material[18].
Data preprocessing
The data were preprocessed by the surface-based rs-fMRI data analysis package DPABISurf (Magnetic Resonance Imaging Research Center, Institute of Psychology, Chinese Academy of Sciences; http://rfmri.org/DPABISurf/) in DPABI on the MATLAB R2020a platform[21]. The default preprocessing pipeline includes field map distortion correction, skull stripping, spatial normalization, brain tissue segmentation, surface reconstruction for T1-weighted images, slice-timing correction, realignment, head-motion estimation, and spatial registration for functional images. The data were processed into two different spaces: FreeSurfer fsaverage5 surface space (for cerebral cortex) and Montreal Neurological Institute volume space (for subcortical nucleus, brain stem, and cerebellum). The seed time series from both the cerebral cortex and subcortical nucleus were further extracted[23,24].
Functional network construction
DPABINet (Magnetic Resonance Imaging Research Center, Institute of Psychology, Chinese Academy of Sciences; http://rfmri.org/DPABINet/) was used to construct a brain functional network. Based on the Schaefer et al[25] and Tian et al[26], the brain was parcellated into 454 regions (400 cortical subregions and 54 subcortical subregions), which represented each node of the functional network, and the brain was further assigned to 7 functional subnetworks defined by Yeo et al[27]. Then, the mean time series for each node was extracted, and the Pearson correlation coefficient between each node pair was computed to evaluate the interregional resting-state FC of each participant. Finally, correlation coefficients were converted to a normal distribution using Fisher’s r-to-z transformation to generate 454 × 454 whole-brain matrices for each subject[28].
Statistical analysis
IBM SPSS 26.0 software was used to compare the between-group differences in demographic (age, gender, years of education) and cognitive function scores (total score of MMSE and memory + recall score) in SA and UA by using the t test or χ2 test for continuous and categorical variables, respectively. The significance level was set at P < 0.05. Analysis of brain network data was conducted using DPABINet software on the MATLAB 2020a platform. The two-sample t test, with gender, age and gray matter density as covariates, was used to compare the network data between the two groups, and a permutation test was performed 5000 times[29]. Then, the statistical results were subjected to network-based statistic (NBS) multiple comparisons correction (edge P < 0.001 and component P < 0.05). In the NBS approach, a cluster (or subnetwork) is defined by the interconnectedness of suprathreshold edges in topological space, providing more statistical power than mass-univariate analysis[30]. Correlation analysis was used to analyze the relationship between functional connections and neuropsychological evaluation among all the subjects (UA group and SA group) with the same covariates and NBS correction. The SVM model was used to construct the model to differentiate SA from UA using R version 4.0.5 and the SVM packages[31]. All differential FCs were used for model development. The performances of the SVM model were shown with the AUC, accuracy, precision, specificity, and sensitivity.
RESULTS
Demographic and neuropsychological evaluation results
The demographic and neuropsychological evaluation data are listed in the Supplementary material and in the previous paper (Supplementary Table 1). Gender, age, and education years were not different between the SA and UA. The total MMSE and memory + recall scores of SA group were significantly higher in comparison to those of UA group.
Differences in brain network analysis results
The BrainNet analysis showed that a total of 17 significantly altered FCs passed multiple comparison correction (NBS correction with edge P < 0.001 and component P < 0.05), including 10 increased FCs and 7 decreased FCs in the SA group compared to the UA group. Detailed information on the 17 edge-based FCs is shown in Table 1. There were 2 increased edge-based FCs with the highest significance (P = 0.0001): One between the left middle frontal gyrus and right occipital gyrus and the other between the right middle frontal gyrus and right angular gyrus (ANG). There were 4 decreased edge-based FCs with the highest significance (P = 0.0005) between the left middle temporal gyrus (MTG) and right posterior central gyrus, right middle occipital gyrus and right ANG, right posterior central gyrus and right superior temporal gyrus, and left middle frontal gyrus and putamen. The most involved networks were the DMN, dorsal attention network (DAN), and limbic network (LN) (Table 1 and Figure 1). In terms of FC nodes, four appeared the most frequently, such as the left middle frontal gyrus, left MTG, right temporal pole: MTG, and right ANG.
Cont-Par3_L (IPL.L) and Limbic-TempPole5_R (TPOmid.R)
3.9510
0.0002
Limbic-OFC4_R (ORBmid.R) and Default-Par4_R (ANG.R)
3.9100
0.0002
SalVentAttn-FrOperIns6_L (INS.L) and Default-PFC14_L (SFGdor.L)
3.8449
0.0003
SalVentAttn-FrOperIns6_L (INS.L) and Default-Temp4_L (MTG.L)
3.8215
0.0003
DorsAttn-Post12_L (IPL.L) and Default-PFCdPFCm3_R (ACG.R)
3.6637
0.0005
DorsAttn-Post12_L (IPL.L) and Limbic-TempPole5_R (TPOmid.R)
3.6497
0.0005
SalVentAttn-PFCl1_L (MFG.L) and Default-PFCdPFCm3_R (ACG.R)
3.5355
0.0007
SomMot28_R (PoCG.R) and HIP-head-m2-lh (HIP.L)
3.5220
0.0008
Vis22_R (MOG.R) and Default-Par4_R (ANG.R)
-3.6689
0.0005
Default-Temp4_L (MTG.L) and SomMot28_R (PoCG.R)
-3.6525
0.0005
DorsAttn-Post13_R (SMG.R) and Limbic-TempPole5_R (TPOmid.R)
-3.6394
0.0005
SalVentAttn-PFCl1_L (MFG.L) and PUT-DP-lh (PUT.R)
-3.6345
0.0005
DorsAttn-Post12_R (SPG.R) and HIP-head-m2-lh (HIP.L)
-3.5673
0.0007
Cont-PFCl11_R (MFG.R) and Default-pCunPCC1_R (PCUN.R)
-3.5402
0.0007
Cont-Par3_L (IPL.L) and Default-Temp4_L (MTG.L)
-3.5058
0.0008
Correlations of FCs and neuropsychological evaluation
The correlation analysis showed that a total of 6 altered FCs significantly correlated with MMSE scores that passed multiple comparison correction (NBS correction with edge P < 0.001 and component P < 0.05), including 5 positively altered FCs and 1 negatively altered FC. The most involved networks were the DMN, DAN, and LN (Table 2 and Figure 2). No other correlations were found between FCs and any demographic data.
Limbic-OFC4_R (ORBmid.R) and Default-Par4_R (ANG.R)
0.4514
0.0001
DorsAttn-Post12_L (IPL.L) and Default-PFCdPFCm3_R (ACG.R)
0.3341
0.0039
Cont-PFCl11_R (MFG.R) and Default-Par4_R (ANG.R)
0.3227
0.0054
DorsAttn-Post12_L (IPL.L) and Limbic-TempPole5_R (TPOmid.R)
0.3047
0.0088
SalVentAttn-FrOperIns6_L (INS.L) and Default-Temp4_L (MTG.L)
0.2915
0.0124
DorsAttn-Post13_R (SMG.R) and Limbic-TempPole5_R (TPOmid.R)
-0.2821
0.0156
SVM model of SA with FC
A total of 17 significantly altered FCs associated with SA were included in the SVM model, and the ROC curve is shown in Figure 3. The area under the curve (AUC) of the SVM model was 0.895 with 95% confidence interval: 0.741-1.000 (DeLong test), and the accuracy (0.909), precision (0.933), sensitivity (0.933), and specificity (0.857) of the SVM model were also determined.
Figure 3 Support vector machine model of successful aging with functional connectivities.
SVM: Support vector machine; ROC: Receiver operating characteristic.
DISCUSSION
Currently, research on SA is relatively rare. We focused on using rs-fMRI to analyze the brain mechanism of aging from a different perspective from aging-related disease. Previous work reported that the brain regions with altered fMRI characteristics (ALFF, ReHo, degree centrality, and voxel-mirrored homotopic connectivity) in the SA group were concentrated in the frontal (6 brain regions) and temporal (4 brain regions) lobes[18]. ALFF in the right orbital part orbital part of the middle frontal gyrus (ORBmid) was significantly correlated with cognitive function and age, which might be used to distinguish SA[18]. The study extends it to explore the potential characteristics of the brain network and FC of SA according to the four-dimensional model of Lu et al[32] in Shanghai. Unlike our previous work, which focused on brain region activity patterns, this study shifts the focus to exploring FC patterns.
The globe brain FC analysis found that in all 17 significantly altered FCs of SA, 6 FCs were correlated with MMSE scores, wherein the 5 FCs of right ORBmid and right ANG, left inferior parietal gyrus (IPL) and right anterior cingulate and paracingulate gyri, left IPL and right temporal pole: MTG (TPOmid), left insula gyrus and left MTG were increased and positively correlated with MMSE scores, but the FC of right supramarginal gyrus (SMG) and right TPOmid was decreased and negatively correlated with MMSE scores. No studies have reported the globe-brain FC of SA. Only Lin et al[33], using fMRI data from the ADNI project and seed-based FC analysis, found that supernomals had significantly stronger FC between the anterior cingulate and paracingulate gyri and R-hippocampus, the median cingulate gyrus (MCG) and left superior temporal gyrus, and the posterior cingulate gyrus and right precuneus and weaker FC between the MCG and right middle frontal gyrus and between the MCG and right thalamus. All of these FCs exhibited notable associations with memory and global cognitive functioning across all participants.
The brain network analysis found that almost all 8 brain networks were involved, and the internetwork FC of 5 networks was related to MMSE performance, of which the DMN, DAN, and LN were the most concentrated networks. The FC of the right ORBmid and right ANG was mostly increased and positively related to the MMSE score, and the FC of the right SMG and right TPOmid was the only one decreased and negatively related to the MMSE score. ORBmid and TPOmid both belong to LN networks. ORBmid participates in many executive functions, such as planning, attention regulation, cognitive flexibility, working memory, behavioral inhibition and emotional regulation[34]. TPOmid plays a number of functional roles in spatial perception, multisensory integration, language processing, attention control and working memory and participates in advanced cognitive processes such as language comprehension, language production and mathematical ability[35]. The ANG is part of the DMN network and is a hub for modulation of the language-related cortex by distinct prefrontal executive control regions[36]. The SMG is part of the DAN network and plays an important role in working memory, executive control, emotion regulation and social cognition[37,38]. Zhang et al[19] found that superagers had stronger connectivity within the DMN and significance network compared to typical older adults, and their connectivity in these networks is similar to that of young adults. Remarkably, the enhanced connectivity within each network served as an independent predictor of superior performance in California Verbal Learning Test and recognition tasks among older adults. It is the only brain network study partly related to SA.
Recently, three activation likelihood estimation meta-analyses with neuroimaging characteristics such as ALFF/fractional amplitude of low-frequency fluctuation, ReHo, and FC reported new progress in DMN and DAN network research in mild cognitive impairment (MCI) and bipolar depression (BD), which could corroborate our results in the opposite direction from the perspective of disease. Disruption of the DMN network is often considered to be a potential biomarker for the progression from MCI to Alzheimer’s disease[39] and is also involved in the pathophysiology of BD[40]. The specific brain region alterations in the DMN network reported by Yuan et al[39] in MCI patients and by Xue et al[40] in BD patients or in the DAN network reported by Wu et al[41] in MCI patients, all mainly occurred in the bilateral/right medial frontal lobe (including ORBmid), bilateral/left inferior parietal lobe (including ANG and SMG), bilateral middle temporal lobe (including TPOmid), etc. The key brain networks and FC nodes mentioned above for MCI and BD were also demonstrated in our results for SA.
Finally, we took a total of 17 significantly altered FCs of SA into the SVM model to explore its differentiation power of SA from UA and found that its AUC was 0.895 with 95% confidence interval: 0.741-1.000, which means that all 17 FCs might be important components of the brain mechanism of SA. This kind of machine learning approach is currently popular. Yang et al[42] also used a logistic regression algorithm to identify chronic insomnia with ALFF and FC features of rs-fMRI and found that the AUC was 0.89, suggesting that these potential neuroimaging biomarkers contribute to the diagnostic identification of chronic insomnia.
This study reveals significant differences in FC features between SA and UA, providing an important foundation for future longitudinal studies. Specifically, changes in the FC of key connection nodes within the DMN, DAN, and LN may reflect how the brain adapts to the challenges of aging. The FC of the right ORBmid in the LN network and the right ANG in the DMN network is mostly increased and positively correlated with MMSE scores, suggesting that SA individuals may maintain cognitive function by enhancing network interactions between these regions. In contrast, the FC of the right SMG in the DAN network and the right TPOmid in the LN network is decreased and negatively correlated with MMSE scores, which may indicate a decline in the brain’s ability to integrate and coordinate cognitive resources during aging, thereby affecting cognitive performance. Understanding these brain network differences that distinguish SA from UA not only serves as an early biomarker for age-related cognitive decline but also helps identify individuals at risk of cognitive impairment before symptoms become apparent, thus providing an opportunity for timely intervention. Additionally, these features provide a basis for developing targeted intervention strategies (such as cognitive training, neurostimulation, etc.), which can help prevent or delay the onset and progression of cognitive decline, thereby offering scientific evidence to promote SA and improve the quality of life in older adults.
This study also has some limitations. First, there is no consensus on the definition of SA, and the small sample size may lead to biased results. Second, to date, only a few studies have been conducted in SA, and the results and their interpretation are still tentative. The rs-fMRI globe-brain FC and brain network studies are based on correlation analysis. The interpretation of our findings needs to be treated with caution to avoid oversimplification or over inference of the causal relationship. Third, due to the cross-sectional design of the study, we cannot observe the dynamic trends of brain networks and FC in SA. Future longitudinal studies are needed to explore the brain functional characteristics of the process of aging.
CONCLUSION
In conclusion, the globe-brain FC analysis and brain network analysis of rs-fMRI research of SA found that the 6 internetwork FCs of 5 brain networks were significantly altered and related to MMSE performance, of which the DMN, DAN, and LN were the most concentrated networks. The FC of the right ORBmid of the LN networks and right ANG of the DMN network was mostly increased and positively related to the MMSE score, and the FC of the right SMG of the DAN network and right TPOmid of the LN network was only decreased and negatively related to the MMSE score. The SVM model with all 17 significantly altered FCs could better differentiate SA individuals. The identification of key brain networks and FC nodes of SA could help us better understand the brain mechanism and further explore neuroimaging biomarkers of SA.
ACKNOWLEDGEMENTS
We thank the subjects and their families who participated in this study and we would like to acknowledge everyone who helped us in this project.
Footnotes
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Psychiatry
Country of origin: China
Peer-review report’s classification
Scientific Quality: Grade C, Grade C
Novelty: Grade B, Grade B
Creativity or Innovation: Grade B, Grade C
Scientific Significance: Grade B, Grade B
P-Reviewer: Zhou Y; Zhao DL S-Editor: Bai Y L-Editor: A P-Editor: Zhao S
Chen Q, Baran TM, Rooks B, O'Banion MK, Mapstone M, Zhang Z, Lin F; Alzheimer's Disease Neuroimaging Initiative. Cognitively supernormal older adults maintain a unique structural connectome that is resistant to Alzheimer's pathology.Neuroimage Clin. 2020;28:102413.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 6][Cited by in RCA: 5][Article Influence: 1.0][Reference Citation Analysis (0)]
Sun FW, Stepanovic MR, Andreano J, Barrett LF, Touroutoglou A, Dickerson BC. Youthful Brains in Older Adults: Preserved Neuroanatomy in the Default Mode and Salience Networks Contributes to Youthful Memory in Superaging.J Neurosci. 2016;36:9659-9668.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 93][Cited by in RCA: 119][Article Influence: 14.9][Reference Citation Analysis (0)]
Gefen T, Peterson M, Papastefan ST, Martersteck A, Whitney K, Rademaker A, Bigio EH, Weintraub S, Rogalski E, Mesulam MM, Geula C. Morphometric and histologic substrates of cingulate integrity in elders with exceptional memory capacity.J Neurosci. 2015;35:1781-1791.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 79][Cited by in RCA: 101][Article Influence: 10.1][Reference Citation Analysis (0)]
Sun J, Zhao X, Zhou J, Dang X, Zhu S, Liu L, Zhou Z. Preliminary Analysis of Volume-Based Resting-State Functional MRI Characteristics of Successful Aging in China.J Alzheimers Dis. 2023;91:767-778.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Yeo BT, Krienen FM, Sepulcre J, Sabuncu MR, Lashkari D, Hollinshead M, Roffman JL, Smoller JW, Zöllei L, Polimeni JR, Fischl B, Liu H, Buckner RL. The organization of the human cerebral cortex estimated by intrinsic functional connectivity.J Neurophysiol. 2011;106:1125-1165.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 6365][Cited by in RCA: 5550][Article Influence: 396.4][Reference Citation Analysis (0)]
Guan XY, Zheng WJ, Fan KY, Han X, Li X, Yan ZH, Lu Z, Gong J. Changes in a sensorimotor network, occipital network, and psychomotor speed within three months after focal surgical injury in pediatric patients with intracranial space-occupying lesions.BMC Pediatr. 2022;22:321.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Park CH, Kim BR, Park HK, Lim SM, Kim E, Jeong JH, Kim GH. Predicting superagers by machine learning classification based on the functional brain connectome using resting-state functional magnetic resonance imaging.Cereb Cortex. 2022;32:4183-4190.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 1][Cited by in RCA: 3][Article Influence: 0.8][Reference Citation Analysis (0)]
Lu Z, Li CB, Zhang MY. [The analysis of stability for discriminating standard on successful aging].Shanghai Jinshen Yixue. 2000;12:187-189, 193.
[PubMed] [DOI][Cited in This Article: ]
Kim H, Wang K, Cutting LE, Willcutt EG, Petrill SA, Leopold DR, Reineberg AE, Thompson LA, Banich MT. The Angular Gyrus as a Hub for Modulation of Language-related Cortex by Distinct Prefrontal Executive Control Regions.J Cogn Neurosci. 2022;34:2275-2296.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 11][Cited by in RCA: 6][Article Influence: 2.0][Reference Citation Analysis (0)]
Şahin MH, Akyüz ME, Karadağ MK, Yalçın A. Supramarginal Gyrus and Angular Gyrus Subcortical Connections: A Microanatomical and Tractographic Study for Neurosurgeons.Brain Sci. 2023;13:430.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Xue C, Zhang X, Cao P, Yuan Q, Liang X, Zhang D, Qi W, Hu J, Xiao C. Evidence of functional abnormalities in the default mode network in bipolar depression: A coordinate-based activation likelihood estimation meta-analysis.J Affect Disord. 2023;326:96-104.
[PubMed] [DOI][Cited in This Article: ][Cited by in RCA: 3][Reference Citation Analysis (0)]
Wu H, Song Y, Chen S, Ge H, Yan Z, Qi W, Yuan Q, Liang X, Lin X, Chen J. An Activation Likelihood Estimation Meta-Analysis of Specific Functional Alterations in Dorsal Attention Network in Mild Cognitive Impairment.Front Neurosci. 2022;16:876568.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 1][Cited by in RCA: 9][Article Influence: 3.0][Reference Citation Analysis (0)]
Yang N, Yuan S, Li C, Xiao W, Xie S, Li L, Jiang G, Ma X. Diagnostic identification of chronic insomnia using ALFF and FC features of resting-state functional MRI and logistic regression approach.Sci Rep. 2023;13:406.
[PubMed] [DOI][Cited in This Article: ][Cited by in RCA: 5][Reference Citation Analysis (0)]
