Exploring memory function in earthquake trauma survivors with resting-state fMRI and machine learning.
Association
Machine learning
Memory
Trauma survivor
fMRI
Journal
BMC psychiatry
ISSN: 1471-244X
Titre abrégé: BMC Psychiatry
Pays: England
ID NLM: 100968559
Informations de publication
Date de publication:
03 02 2020
03 02 2020
Historique:
received:
05
03
2019
accepted:
21
01
2020
entrez:
5
2
2020
pubmed:
6
2
2020
medline:
11
11
2020
Statut:
epublish
Résumé
Traumatized earthquake survivors may develop poor memory function. Resting-state functional magnetic resonance imaging (rs-fMRI) and machine learning techniques may one day aid the clinical assessment of individual psychiatric patients. This study aims to use machine learning with Rs-fMRI from the perspectives of neurophysiology and neuroimaging to explore the association between it and the individual memory function of trauma survivors. Rs-fMRI data was acquired for eighty-nine survivors (male (33%), average age (SD):45.18(6.31) years) of Wenchuan earthquakes in 2008 each of whom was screened by experienced psychiatrists based on the clinician-administered post-traumatic stress disorder (PTSD) scale (CAPS), and their memory function scores were determined by the Wechsler Memory Scale-IV (WMS-IV). We explored which memory function scores were significantly associated with CAPS scores. Using simple multiple kernel learning (MKL), Rs-fMRI was used to predict the memory function scores that were associated with CAPS scores. A support vector machine (SVM) was also used to make classifications in trauma survivors with or without PTSD. Spatial addition (SA), which is defined by spatial working memory function, was negatively correlated with the total CAPS score (r = - 0.22, P = 0.04). The use of simple MKL allowed quantitative association of SA scores with statistically significant accuracy (correlation = 0.28, P = 0.03; mean squared error = 8.36; P = 0.04). The left middle frontal gyrus and the left precuneus contributed the largest proportion to the simple MKL association frame. The SVM could not make a quantitative classification of diagnosis with statistically significant accuracy. The use of the cross-sectional study design after exposure to an earthquake and the leave-one-out cross-validation (LOOCV) increases the risk of overfitting. Spontaneous brain activity of the left middle frontal gyrus and the left precuneus acquired by rs-fMRI may be a brain mechanism of visual working memory that is related to PTSD symptoms. Machine learning may be a useful tool in the identification of brain mechanisms of memory impairment in trauma survivors.
Sections du résumé
BACKGROUND
Traumatized earthquake survivors may develop poor memory function. Resting-state functional magnetic resonance imaging (rs-fMRI) and machine learning techniques may one day aid the clinical assessment of individual psychiatric patients. This study aims to use machine learning with Rs-fMRI from the perspectives of neurophysiology and neuroimaging to explore the association between it and the individual memory function of trauma survivors.
METHODS
Rs-fMRI data was acquired for eighty-nine survivors (male (33%), average age (SD):45.18(6.31) years) of Wenchuan earthquakes in 2008 each of whom was screened by experienced psychiatrists based on the clinician-administered post-traumatic stress disorder (PTSD) scale (CAPS), and their memory function scores were determined by the Wechsler Memory Scale-IV (WMS-IV). We explored which memory function scores were significantly associated with CAPS scores. Using simple multiple kernel learning (MKL), Rs-fMRI was used to predict the memory function scores that were associated with CAPS scores. A support vector machine (SVM) was also used to make classifications in trauma survivors with or without PTSD.
RESULTS
Spatial addition (SA), which is defined by spatial working memory function, was negatively correlated with the total CAPS score (r = - 0.22, P = 0.04). The use of simple MKL allowed quantitative association of SA scores with statistically significant accuracy (correlation = 0.28, P = 0.03; mean squared error = 8.36; P = 0.04). The left middle frontal gyrus and the left precuneus contributed the largest proportion to the simple MKL association frame. The SVM could not make a quantitative classification of diagnosis with statistically significant accuracy.
LIMITATIONS
The use of the cross-sectional study design after exposure to an earthquake and the leave-one-out cross-validation (LOOCV) increases the risk of overfitting.
CONCLUSION
Spontaneous brain activity of the left middle frontal gyrus and the left precuneus acquired by rs-fMRI may be a brain mechanism of visual working memory that is related to PTSD symptoms. Machine learning may be a useful tool in the identification of brain mechanisms of memory impairment in trauma survivors.
Identifiants
pubmed: 32013935
doi: 10.1186/s12888-020-2452-5
pii: 10.1186/s12888-020-2452-5
pmc: PMC6998246
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
43Subventions
Organisme : National Key Research & Development Program of China
ID : 2016YFC1307200
Pays : International
Organisme : Major Research Plan
ID : 81371484, 81701328, 81371527, 81030027 and 81220108013
Pays : International
Organisme : University Natural Science Research Project of Anhui Province (CN)
ID : 2016YFC1307201
Pays : International
Organisme : Postdoctoral Research Foundation of China
ID : 2017M612972
Pays : International
Organisme : the Support Plan of Sichuan,China
ID : 2014GZ0143
Pays : International
Références
Psychother Psychosom. 2016;85(1):16-26
pubmed: 26610167
Neuroimage. 2011 Aug 1;57(3):1221-33
pubmed: 21609772
Brain Cogn. 2013 Feb;81(1):151-9
pubmed: 23207576
Physiol Behav. 2006 Jan 30;87(1):45-50
pubmed: 16182325
Psychol Aging. 2013 Jun;28(2):304-13
pubmed: 23421323
Psychiatry Investig. 2017 Jan;14(1):21-29
pubmed: 28096871
Psychiatry Res. 2013 Sep 30;213(3):249-55
pubmed: 23830931
Neuroimage. 2014 Feb 15;87:345-55
pubmed: 24220041
Am J Psychiatry. 2010 Jul;167(7):748-51
pubmed: 20595427
Neuropsychopharmacology. 2014 Feb;39(3):681-7
pubmed: 24064470
Biol Psychiatry. 2003 Mar 15;53(6):474-81
pubmed: 12644352
Neuroinformatics. 2016 Jul;14(3):339-51
pubmed: 27075850
Biol Psychiatry. 2002 Dec 1;52(11):1079-88
pubmed: 12460691
J Trauma Stress. 2005 Dec;18(6):617-29
pubmed: 16382423
J Neurosci Methods. 2008 Jul 15;172(1):137-41
pubmed: 18501969
J Affect Disord. 2015 Sep 01;183:263-8
pubmed: 26042728
Neuroimage. 2014 Jan 1;84:466-75
pubmed: 24045077
J Trauma Stress. 2007 Aug;20(4):435-47
pubmed: 17721957
Lifetime Data Anal. 2011 Jul;17(3):461-72
pubmed: 21140212
Arch Gen Psychiatry. 1998 Jul;55(7):626-32
pubmed: 9672053
JAMA. 1988 May 13;259(18):2725-9
pubmed: 3282086
Dialogues Clin Neurosci. 2009;11(4):389-96
pubmed: 20135896
Science. 2010 Sep 10;329(5997):1358-61
pubmed: 20829489
J Trauma Stress. 1995 Oct;8(4):505-25
pubmed: 8564271
Psychol Bull. 2014 Jan;140(1):69-97
pubmed: 23914721
Ann N Y Acad Sci. 1997 Jun 21;821:208-18
pubmed: 9238205
Ann N Y Acad Sci. 2006 Jul;1071:54-66
pubmed: 16891562
Bioinformatics. 2004 Nov 1;20(16):2626-35
pubmed: 15130933
Nature. 1993 Jun 17;363(6430):623-5
pubmed: 8510752
J Clin Psychol. 2002 Jan;58(1):23-41
pubmed: 11748595
J Affect Disord. 2002 Jun;70(1):1-17
pubmed: 12113915
Neuroscience. 2006 Apr 28;139(1):339-49
pubmed: 16324793
Proc Natl Acad Sci U S A. 2009 Sep 8;106(36):15412-7
pubmed: 19720989
Trends Psychiatry Psychother. 2017 Apr-Jun;39(2):135-143
pubmed: 28700042
Brain Imaging Behav. 2016 Jun;10(2):387-407
pubmed: 26138580
Proc Natl Acad Sci U S A. 1996 Nov 26;93(24):13473-80
pubmed: 8942959
J Abnorm Psychol. 2001 Feb;110(1):40-8
pubmed: 11261398
Neuropsychobiology. 2012;66(3):174-84
pubmed: 22948482
Hippocampus. 2012 May;22(5):1154-63
pubmed: 21739522
Behav Res Ther. 2001 Apr;39(4):373-93
pubmed: 11280338
Psychol Med. 2014 Jan;44(1):195-203
pubmed: 23551879
Neuropsychology. 2002 Jul;16(3):370-9
pubmed: 12146684
Nord J Psychiatry. 2003;57(3):191-8
pubmed: 12775293
Am J Psychiatry. 2006 Apr;163(4):652-8
pubmed: 16585440
J Psychiatr Res. 2011 May;45(5):650-9
pubmed: 21106206
Clin Psychol Rev. 2000 Nov;20(8):1041-65
pubmed: 11098399
J Psychiatr Res. 2008 Jul;42(8):659-69
pubmed: 17698081
Memory. 2012;20(3):277-99
pubmed: 22348400
J Psychiatry Neurosci. 2013 Nov;38(6):381-7
pubmed: 23710694
J Learn Disabil. 2014 Jul-Aug;47(4):349-65
pubmed: 23124381
Biol Psychiatry. 2015 Aug 15;78(4):278-86
pubmed: 25702259
Am J Psychiatry. 1991 Jan;148(1):10-20
pubmed: 1824611
Psychol Med. 2012 May;42(5):1037-47
pubmed: 22059690
Neuron. 2011 Nov 17;72(4):665-78
pubmed: 22099467
Psychiatry Res. 2002 Jul 31;110(3):231-8
pubmed: 12127473
Compr Psychiatry. 2014 Oct;55(7):1688-95
pubmed: 25060989
Eur J Psychotraumatol. 2017 Feb 14;8(1):1283086
pubmed: 28326164
Neuroimage. 2012 Feb 1;59(3):2142-54
pubmed: 22019881
J Nerv Ment Dis. 1976 Oct;163(4):246-52
pubmed: 972326
Eur Child Adolesc Psychiatry. 2013 Dec;22(12):745-55
pubmed: 23553572
Front Neurosci. 2016 Jun 24;10:292
pubmed: 27445664
Front Aging Neurosci. 2015 Aug 31;7:169
pubmed: 26379547
Cereb Cortex. 1996 Jan-Feb;6(1):31-8
pubmed: 8670636
Hum Mov Sci. 2005 Oct-Dec;24(5-6):644-56
pubmed: 16337295
J Trauma Stress. 2014 Oct;27(5):576-84
pubmed: 25322887
J Anxiety Disord. 2013 May;27(4):413-9
pubmed: 23746489
Magn Reson Med. 1996 Mar;35(3):346-55
pubmed: 8699946
Neuropsychopharmacology. 2016 Jan;41(2):611-8
pubmed: 26111649
PLoS One. 2016 Feb 10;11(2):e0144766
pubmed: 26863536
Am Psychol. 1990 Sep;45(9):1043-56
pubmed: 2221571