Share：
Share this content in WeChat
X
[Chinese][PDF]225577
• Technical Article •
Features fusion of brain networks and its application to autism recognition by machine learning based on resting-state functional magnetic resonance imaging
ZHAO Xiaohu  GE Manling  CHEN Shenghua  WANG Lei  SONG Zibo  XIE Chong  YANG Zekun 

Cite this article as: Zhao XH, Ge ML, Chen SH, et al. Features fusion of brain networks and its application to autism recognition by machine learning based on resting-state functional magnetic resonance imaging[J]. Chin J Magn Reson Imaging, 2021, 12(12): 55-61. DOI:10.12015/issn.1674-8034.2021.12.011.


[Abstract] Objective The resting-state functional magnetic resonance imaging (rs-fMRI) technology combined with machine learning algorithm was used to classify the patients with autism, trying to provide reference for early identification of autism. Materials and Methods: The rs-fMRI data of a total of 24 patients with autism and 25 healthy people were pre-processed. Then, the partial correlation functional connection (FC) was used to construct the network and the sparsity space of 0.05—0.50 with a step size of 0.05, the brain functional networks were constructed by GRETNA software.A total of 4 local nodes metrics were calculated respectively for patients and healthy individuals. Finally, the proportion of classification accuracy of each index is used as the weight coefficient for feature fusion, so as to construct the feature vector, input into the support vector machine model for classification and cross validation to test the feature fusion effect.Results The average accuracy of weighted features fusion can reach up 89.47%, which is 21.05% higher than that of a single feature and 4.74% higher than non-weighted feature fusion method.Conclusions This work might provide a new index and a new method to recognize the autism by rs-fMRI.
[Keywords] autism spectrum disorders；brain functional network；weighted feature fusion；machine learning；resting-state functional magnetic resonance imaging

ZHAO Xiaohu1, 2   GE Manling1, 2   CHEN Shenghua1, 2*   WANG Lei3*   SONG Zibo1, 2   XIE Chong1, 2   YANG Zekun1, 2  

1 State Key Laboratory of Reliability and Intelligence of Electrical Equipment, Hebei University of Technology, Tianjin 300130, China

2 Key Laboratory of Electromagnetic Field and Electrical Apparatus Reliability of Hebei Province, Hebei University of Technology, Tianjin 300130, China

3 Langfang Polytechnic College, Langfang 065001, China

Chen SH, E-mail: chenshenghua@hebut.edu.cn Wang L, E-mail: wanglei1982615@163.com

Conflicts of interest   None.

ACKNOWLEDGMENTS Natural Science Foundation of Hebei Province (No. E2019202019); Key Projects of Science and Technology Research in Hebei Province (No. ZD2021025).
Received  2021-07-17
Accepted  2021-09-18
DOI: 10.12015/issn.1674-8034.2021.12.011
Cite this article as: Zhao XH, Ge ML, Chen SH, et al. Features fusion of brain networks and its application to autism recognition by machine learning based on resting-state functional magnetic resonance imaging[J]. Chin J Magn Reson Imaging, 2021, 12(12): 55-61. DOI:10.12015/issn.1674-8034.2021.12.011.

[1]
Pua EPK, Bowden SC, Seal ML. Autism spectrum disorders: Neuroimaging findings from systematic reviews[J]. Research in Autism Spectrum Disorders, 2017, 34: 28-33. DOI: 10.1016/j.rasd.2016.11.005.
[2]
Brentani H, Paula CSD, Bordini D, et al. Autism spectrum disorders: an overview on diagnosis and treatment[J]. Brazilian Journal of Psychiatry, 2013, 35: S62-S72. DOI: 10.1590/1516-4446-2013-S104.
[3]
Prelock PA, Disorders S. Autism spectrum disorders[J]. The Handbook of Language, 2021: 129-151. DOI: 10.1002/9781119606987.ch7.
[4]
Hodges H, Fealko C, Soares N. Autism spectrum disorder: definition, epidemiology, causes, and clinical evaluation[J]. Translational Pediatrics, 2020, 9(Suppl 1): S55-S65. DOI: 10.21037/tp.2019.09.09.
[5]
Pierce K, Courchesne E, Bacon E. To screen or not to screen universally for autism is not the question: Why the task force got it wrong[J]. The Journal of Pediatrics, 2016, 176: 182-194. DOI: 10.1016/j.jpeds.2016.06.004.
[6]
Liao MY, Chen LY, Wang GS, et al. Intelligent recognition of children with autism spectrum disorder based on multi-modal data and its effectiveness[J]. Chinese Science Bulletin, 2021, 66(20): 2618-2628.
[7]
Zeng K, Kang JN, Ouyang GX, et al. Disrupted brain network in children with autism spectrum disorder[J]. Scientific Reports, 2017, 7(1): 1-12. DOI: 10.1038/s41598-017-16440-z.
[8]
Weng SJ, Wiggins JL, Peltier SJ, et al. Alterations of resting state functional connectivity in the default network in adolescents with autism spectrum disorders[J]. Brain Research, 2010, 1313: 202-214. DOI: 10.1016/j.brainres.2009.11.057.
[9]
Liu J, Li M, Pan Y, et al. Complex brain network analysis and its applications to brain disorders: a survey[J]. Complexity, 2017, 2017. DOI: 10.1155/2017/8362741.
[10]
Kocevar G, Stamile C, Hannoun S, et al. Graph theory-based brain connectivity for automatic classification of multiple sclerosis clinical courses[J]. Frontiers in Neuroscience, 2016, 10: 478. DOI: 10.3389/fnins.2016.00478.
[11]
Sacchet MD, Prasad G, Foland-Ross LC, et al. Support vector machine classification of major depressive disorder using diffusion-weighted neuroimaging and graph theory[J]. Frontiers in psychiatry, 2015, 6: 21. DOI: 10.3389/fpsyt.2015.00021.
[12]
Fu CY, Wang Y, Xiao HB, et al. Classification of depression using fusion features based on multi-scale functional brain network[J]. Chinese Journal of Medical Physics, 2020, 37(4): 439-444. DOI: 1005-202X(2020)04-0439-06.
[13]
Zhu YQ, Wang WG. Research on Multi-dimensional Feature Extraction and Classification Algorithm of EEG Signal[J/OL]. Computer Engineering and Applications, 2021,1-10. http://kns.cnki.net/kcms/detail/11.2127.TP.20210414.1010.006.html.
[14]
Yan CG, Zang YF. DPARSF: a MATLAB toolbox for" pipeline" data analysis of resting-state fMRI[J]. Frontiers in Systems Neuroscience, 2010, 4: 13. DOI: 10.3389/fnsys.2010.00013.
[15]
Tzourio-Mazoyer N, Landeau 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. DOI: 10.1006/nimg.2001.0978.
[16]
Huang XX, He HX. Analysis of brain function binding based on partial correlation coefficient[J]. Computer Applications and Software, 2017, 34(09): 103107. DOI: 10.3969/j.issn.1000386x.2017.09.021.
[17]
Ran Q, Jamoulle T, Schaeverbeke J, et al. Reproducibility of graph measures at the subject level using resting‐state fMRI[J]. Brain and Behavior, 2020, 10(8): 2336-2351. DOI: 10.1002/brb3.1705.
[18]
Zhang TT, Zhao ZZ, Zhang C, et al. Classification of early and late mild cognitive impairment using functional brain network of resting-state fMRI[J]. Frontiers in Psychiatry, 2019, 10: 572. DOI: 10.3389/fpsyt.2019.00572.
[19]
Wang JH, Wang XD, Xia MR, et al. GRETNA: a graph theoretical network analysis toolbox for imaging connectomics[J]. Frontiers in human neuroscience, 2015, 9: 386. DOI: 10.3389/fnhum.2015.00386.
[20]
Chen Y, Huang XQ, Wu M, et al. Disrupted brain functional networks in drug-naïve children with attention deficit hyperactivity disorder assessed using graph theory analysis[J]. Human Brain Mapping, 2019, 40(17): 4877-4887. DOI: 10.1002/hbm.24743.
[21]
Xu XW, Li WK, Tao ML, et al. Effective and accurate diagnosis of subjective cognitive decline based on functional connection and graph theory view[J]. Frontiers in neuroscience, 2020, 14: 577887. DOI: 10.3389/fnins.2020.577887.
[22]
Fan FM, Liao XH, Lei TY, et al. Development of the default-mode network during childhood and adolescence: A longitudinal resting-state fMRI study[J]. NeuroImage, 2021, 226: 117581. DOI: 10.1016/j.neuroimage.2020.117581.
[23]
Fang H, Wu QR, Li Y, et al. Structural networks in children with autism spectrum disorder with regression: A graph theory study[J]. Behavioural Brain Research, 2020, 378: 112262. DOI: 10.1016/j.bbr.2019.112262.
[24]
Saqlain SM, Sher M, Shah FA, et al. Fisher score and Matthews correlation coefficient-based feature subset selection for heart disease diagnosis using support vector machines[J]. Knowledge and Information Systems, 2019, 58(1): 139-167. DOI: 10.1007/s10115-018-1185-y.
[25]
Rakic M, Cabezas M, Kushibar K, et al. Improving the detection of autism spectrum disorder by combining structural and functional MRI information[J]. NeuroImage: Clinical, 2020, 25: 102181. DOI: 10.1016/j.nicl.2020.102181.
[26]
Morgado PM, Silveira M. Minimal neighborhood redundancy maximal relevance: Application to the diagnosis of Alzheimer׳s disease[J]. Neurocomputing, 2015, 155: 295-308. DOI: 10.1016/j.neucom.2014.12.070.
[27]
Kang L. Research and application of brain mechanism based on multi-feature fusion[D]. University of Electronic Science and Technology of China, 2020. DOI: 10.27005/d.cnki.gdzku.2020.003235.
[28]
Jie B, Zhang DQ, Gao W, et al. Integration of network topological and connectivity properties for neuroimaging classification[J]. IEEE Transactions on Biomedical Engineering, 2013, 61(2): 576-589. DOI: 10.1109/TBME.2013.2284195.
[29]
Li Y, Liu JY, Tang ZY, et al. Deep spatial-temporal feature fusionfrom adaptive dynamic functional connectivity for MCI Identification[J]. IEEE Transactions on Medical Imaging, 2020, 39(9): 2818-2830. DOI: 10.1109/TMI.2020.2976825.
[30]
Leech R, Sharp DJ. The role of the posterior cingulate cortex in cognition and disease[J]. Brain, 2014, 137(1): 12-32. DOI: 10.1093/brain/awt162.
[31]
Zanghieri M, Menichetti G, Retico A, et al. Node centrality measures identify relevant structural MRI features of subjects with autism[J]. Brain Sciences, 2021, 11(4): 498. DOI: 10.3390/brainsci11040498.
[32]
Hashem S, Nisar S, Bhat AA, et al. Genetics of structural and functional brain changes in autism spectrum disorder[J]. Translational Psychiatry, 2020, 10(1): 229. DOI: 10.1038/s41398-020-00921-3.
[33]
Liu Y, Zhu ZJ, Cao RM, et al. Early brain overgrowth in children with autism spectrum disorders based on structural magnetic resonance imaging[J]. Chin J Magn Reson Imaging, 2020, 11(4): 264-269. DOI: 10.12015/issn.1674-8034.2020.04.005.
[34]
Rudie JD, Shehzad Z, Hernandez LM, et al. Reduced functional integration and segregation of distributed neural systems underlying social and emotional information processing in autism spectrum disorders[J]. Cerebral Cortex, 2012, 22(5): 1025-1037. DOI: 10.1093/cercor/bhr171.
[35]
Hernandez LM, Rudie JD, Green SA, et al. Neural signatures of autism spectrum disorders: insights into bain network dynamics[J]. Neuropsychopharmacology, 2015, 40(1): 171-189. DOI: 10.1038/npp.2014.172.
[36]
He C, Cortes JM, Kang X, et al. Individual‐based morphological brain network organization and its association with autistic symptoms in young children with autism spectrum disorder[J]. Human Brain Mapping, 2021, 42(10): 3282-3294. DOI: 10.1002/hbm.25434.
[37]
Weng SJ, Wiggins JL, Peltier SJ, et al. Alterations of resting state functional connectivity in the default network in adolescents with autism spectrum disorders[J]. Brain Research, 2010, 1313: 202-214. DOI: 10.1016/j.brainres.2009.11.057.
[38]
Korhonen O, Saarimäki H, Glerean E, et al. Consistency of regions of interest as nodes of fMRI functional brain networks[J]. Network Neuroscience, 2017, 1(3): 254-274. DOI: 10.1162/NETN_a_00013.

PREV Feasibility study on quantification method of fat infiltration in thigh muscle based on axial T1WI image
NEXT The effect of fast scan parameters on evaluation of T1, T2 relaxation times and proton density of normal brain using synthetic MRI
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn