White Matter Connectome Edge Density in Children with Autism Spectrum Disorders: Potential Imaging Biomarkers Using Machine-Learning Models.

Item request has been placed! ×
Item request cannot be made. ×
loading   Processing Request
  • Additional Information
    • Source:
      Publisher: Mary Ann Liebert, Inc Country of Publication: United States NLM ID: 101550313 Publication Model: Print Cited Medium: Internet ISSN: 2158-0022 (Electronic) Linking ISSN: 21580014 NLM ISO Abbreviation: Brain Connect
    • Publication Information:
      Original Publication: New Rochelle, NY : Mary Ann Liebert, Inc.
    • Subject Terms:
    • Abstract:
      Prior neuroimaging studies have reported white matter network underconnectivity as a potential mechanism for autism spectrum disorder (ASD). In this study, we examined the structural connectome of children with ASD using edge density imaging (EDI), and then applied machine-learning algorithms to identify children with ASD based on tract-based connectivity metrics. Boys aged 8-12 years were included: 14 with ASD and 33 typically developing children. The edge density (ED) maps were computed from probabilistic streamline tractography applied to high angular resolution diffusion imaging. Tract-based spatial statistics was used for voxel-wise comparison and coregistration of ED maps in addition to conventional diffusion tensor imaging (DTI) metrics of fractional anisotropy (FA), mean diffusivity (MD), and radial diffusivity (RD). Tract-based average DTI/connectome metrics were calculated and used as input for different machine-learning models: naïve Bayes, random forest, support vector machines (SVMs), and neural networks. For these models, cross-validation was performed with stratified random sampling ( × 1,000 permutations). The average accuracy among validation samples was calculated. In voxel-wise analysis, the body and splenium of corpus callosum, bilateral superior and posterior corona radiata, and left superior longitudinal fasciculus showed significantly lower ED in children with ASD; whereas, we could not find significant difference in FA, MD, and RD maps between the two study groups. Overall, machine-learning models using tract-based ED metrics had better performance in identification of children with ASD compared with those using FA, MD, and RD. The EDI-based random forest models had greater average accuracy (75.3%), specificity (97.0%), and positive predictive value (81.5%), whereas EDI-based polynomial SVM had greater sensitivity (51.4%) and negative predictive values (77.7%). In conclusion, we found reduced density of connectome edges in the posterior white matter tracts of children with ASD, and demonstrated the feasibility of connectome-based machine-learning algorithms in identification of children with ASD.
    • References:
      J Int Neuropsychol Soc. 2012 May;18(3):521-9. (PMID: 22390821)
      PLoS Comput Biol. 2017 Jun 22;13(6):e1005550. (PMID: 28640803)
      Autism Res. 2012 Oct;5(5):289-313. (PMID: 22786754)
      Neuroimage Clin. 2013 Jun 23;2:844-53. (PMID: 24179836)
      Psychiatry Res Neuroimaging. 2017 May 30;263:127-134. (PMID: 28384486)
      Neuroimage. 2015 Nov 15;122:318-31. (PMID: 26260428)
      JAMA Psychiatry. 2017 Nov 1;74(11):1120-1128. (PMID: 28877317)
      Brain. 2005 Feb;128(Pt 2):268-76. (PMID: 15548557)
      J Dev Behav Pediatr. 2006 Apr;27(2 Suppl):S95-S103. (PMID: 16685191)
      Front Hum Neurosci. 2016 May 11;10:211. (PMID: 27242476)
      Neuroimage. 2016 Jan 1;124(Pt B):1108-1114. (PMID: 26364861)
      Neurosci Bull. 2017 Apr;33(2):219-237. (PMID: 28283808)
      Neuroimage. 2006 Sep;32(3):989-94. (PMID: 16854598)
      Hum Brain Mapp. 2016 Aug;37(8):2833-48. (PMID: 27219475)
      Radiology. 2016 Dec;281(3):665-668. (PMID: 27870619)
      AMIA Annu Symp Proc. 2007 Oct 11;:686-90. (PMID: 18693924)
      J Autism Dev Disord. 2000 Jun;30(3):205-23. (PMID: 11055457)
      Natl Health Stat Report. 2013 Mar 20;(65):1-11, 1 p following 11. (PMID: 24988818)
      AJNR Am J Neuroradiol. 2012 Dec;33(11):2123-8. (PMID: 22653326)
      Neuroimage. 2018 Jun;173:592-603. (PMID: 29407457)
      Brain Connect. 2016 Jun;6(5):415-33. (PMID: 27021440)
      Prog Neuropsychopharmacol Biol Psychiatry. 2007 Jan 30;31(1):41-5. (PMID: 16863674)
      Prog Neuropsychopharmacol Biol Psychiatry. 2018 Mar 2;82:242-248. (PMID: 29128446)
      J Am Acad Child Adolesc Psychiatry. 2004 Mar;43(3):349-57. (PMID: 15076269)
      Neuroimage. 2006 Jul 1;31(3):968-80. (PMID: 16530430)
      Pediatr Res. 2011 May;69(5 Pt 2):63R-8R. (PMID: 21289538)
      Harv Rev Psychiatry. 2015 Jul-Aug;23(4):223-44. (PMID: 26146755)
      Front Neuroanat. 2016 Jan 26;9:169. (PMID: 26858611)
      Handb Clin Neurol. 2018;147:321-329. (PMID: 29325621)
      Neuroimage. 2004 Sep;23(1):364-9. (PMID: 15325384)
      Neuropsychologia. 2011 Apr;49(5):1378-1382. (PMID: 21333661)
      Neuroimage. 2015 Apr 1;109:402-17. (PMID: 25592996)
      PLoS One. 2014 Jul 30;9(7):e103038. (PMID: 25075609)
      Brain Connect. 2016 Jun;6(5):403-14. (PMID: 26973154)
      Hum Brain Mapp. 2014 Apr;35(4):1284-96. (PMID: 23334984)
      J Autism Dev Disord. 1994 Oct;24(5):659-85. (PMID: 7814313)
      Hum Brain Mapp. 2014 Dec;35(12):5861-76. (PMID: 25044213)
      Neuroimage. 2017 Apr 15;150:162-176. (PMID: 28188913)
      Brain Connect. 2016 Nov;6(9):691-699. (PMID: 27555361)
      Res Dev Disabil. 2014 Nov;35(11):2921-32. (PMID: 25108609)
      Brain Connect. 2016 Sep;6(7):548-57. (PMID: 27345586)
    • Grant Information:
      K23 MH083890 United States MH NIMH NIH HHS
    • Contributed Indexing:
      Keywords: autism; diffusion tensor imaging; edge density imaging; machine learning
    • Accession Number:
      0 (Biomarkers)
    • Publication Date:
      Date Created: 20190122 Date Completed: 20190806 Latest Revision: 20240214
    • Publication Date:
      20240214
    • Accession Number:
      PMC6444925
    • Accession Number:
      10.1089/brain.2018.0658
    • Accession Number:
      30661372