Cognitive Challenges Are Better in Distinguishing Binge From Nonbinge Drinkers: An Exploratory Deep-Learning Study of fMRI Data of Multiple Behavioral Tasks and Resting State.
alcohol use disorder
binge drinking
gamble
interpretability
neural network
Journal
Journal of magnetic resonance imaging : JMRI
ISSN: 1522-2586
Titre abrégé: J Magn Reson Imaging
Pays: United States
ID NLM: 9105850
Informations de publication
Date de publication:
03 2023
03 2023
Historique:
revised:
13
06
2022
received:
08
12
2021
accepted:
16
06
2022
pubmed:
10
7
2022
medline:
22
2
2023
entrez:
9
7
2022
Statut:
ppublish
Résumé
Studies have identified imaging markers of binge drinking. Functional connectivity during both task challenges and resting state was shown to distinguish binge and nonbinge drinkers. However, no studies have compared the efficacy of task and resting data in the classification. Task outperforms resting-state functional magnetic resonance imaging (fMRI) data in the differentiation of binge and nonbinge drinkers. We tested the hypothesis via multiple deep learning algorithms. Cross-sectional; retrospective. A total of 149 binge (107 men) and 151 demographically matched, nonbinge (92 men) drinkers curated from the Human Connectome Project, with 80% randomly selected for model development and 20% for validation/test. A 3 T; fMRI with a blood oxygen level-dependent (BOLD) gradient-echo echo-planar sequence. FMRI data of resting state and seven behavioral tasks were acquired. Graph convolutional network (GCN), long short-term memory, convolutional, and recurrent neural network models were built to distinguish bingers and nonbingers using connectivity matrices of 8, 116, and 268 regions of interest (ROI). Nodal metrics including betweenness centrality, degree centrality, clustering coefficient, efficiency, local efficiency, and shortest path length were calculated from the GCN model. Model performance was quantified by the area under the curve (AUC) in receiver operating characteristic analysis. A P value < 0.05 was considered statistically significant. Task outperformed resting data in classification by approximately 8% by AUC in the test set. Across models and ROI sets, the gambling, motor, language and working memory tasks, each with AUC of 0.614, 0.612, 0.605, and 0.603, performed better than resting data (AUC = 0.548). Models with 116 ROIs (AUC = 0.602) consistently outperformed those with 8 ROIs (AUC = 0.569). Task data performed best with GCN (AUC = 0.619). Nodal metrics of left supplementary motor area and right cuneus showed significant group main effect across tasks. Neural responses to cognitive challenges relative to resting state better characterize binge drinking. The performance of different network models may depend on behavioral tasks and the number of ROIs. 3 TECHNICAL EFFICACY: Stage 2.
Sections du résumé
BACKGROUND
Studies have identified imaging markers of binge drinking. Functional connectivity during both task challenges and resting state was shown to distinguish binge and nonbinge drinkers. However, no studies have compared the efficacy of task and resting data in the classification.
HYPOTHESIS
Task outperforms resting-state functional magnetic resonance imaging (fMRI) data in the differentiation of binge and nonbinge drinkers. We tested the hypothesis via multiple deep learning algorithms.
STUDY TYPE
Cross-sectional; retrospective.
POPULATION
A total of 149 binge (107 men) and 151 demographically matched, nonbinge (92 men) drinkers curated from the Human Connectome Project, with 80% randomly selected for model development and 20% for validation/test.
FIELD STRENGTH/SEQUENCE
A 3 T; fMRI with a blood oxygen level-dependent (BOLD) gradient-echo echo-planar sequence.
ASSESSMENT
FMRI data of resting state and seven behavioral tasks were acquired. Graph convolutional network (GCN), long short-term memory, convolutional, and recurrent neural network models were built to distinguish bingers and nonbingers using connectivity matrices of 8, 116, and 268 regions of interest (ROI). Nodal metrics including betweenness centrality, degree centrality, clustering coefficient, efficiency, local efficiency, and shortest path length were calculated from the GCN model.
STATISTICAL TESTS
Model performance was quantified by the area under the curve (AUC) in receiver operating characteristic analysis. A P value < 0.05 was considered statistically significant.
RESULTS
Task outperformed resting data in classification by approximately 8% by AUC in the test set. Across models and ROI sets, the gambling, motor, language and working memory tasks, each with AUC of 0.614, 0.612, 0.605, and 0.603, performed better than resting data (AUC = 0.548). Models with 116 ROIs (AUC = 0.602) consistently outperformed those with 8 ROIs (AUC = 0.569). Task data performed best with GCN (AUC = 0.619). Nodal metrics of left supplementary motor area and right cuneus showed significant group main effect across tasks.
CONCLUSION
Neural responses to cognitive challenges relative to resting state better characterize binge drinking. The performance of different network models may depend on behavioral tasks and the number of ROIs.
EVIDENCE LEVEL
3 TECHNICAL EFFICACY: Stage 2.
Substances chimiques
Ethanol
3K9958V90M
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
856-868Subventions
Organisme : NIDA NIH HHS
ID : DA051922
Pays : United States
Organisme : NIMH NIH HHS
ID : U54 MH091657
Pays : United States
Commentaires et corrections
Type : CommentIn
Informations de copyright
© 2022 International Society for Magnetic Resonance in Medicine.
Références
Ryan SA, Kokotailo P, Committee on Substance USE, Prevention. Alcohol use by youth. Pediatrics 2019;144(1):e20191357.
Spear LP. Effects of adolescent alcohol consumption on the brain and behaviour. Nat Rev Neurosci 2018;19(4):197-214.
Li G, Chen Y, Le TM, et al. Perceived friendship and binge drinking in young adults: A study of the human connectome project data. Drug Alcohol Depend 2021;224:108731.
Morales AM, Jones SA, Ehlers A, Lavine JB, Nagel BJ. Ventral striatal response during decision making involving risk and reward is associated with future binge drinking in adolescents. Neuropsychopharmacology 2018;43(9):1884-1890.
Courtney AL, Rapuano KM, Sargent JD, Heatherton TF, Kelley WM. Reward system activation in response to alcohol advertisements predicts college drinking. J Stud Alcohol Drugs 2018;79(1):29-38.
Nikolova YS, Knodt AR, Radtke SR, Hariri AR. Divergent responses of the amygdala and ventral striatum predict stress-related problem drinking in young adults: Possible differential markers of affective and impulsive pathways of risk for alcohol use disorder. Mol Psychiatry 2016;21(3):348-356.
Shen X, Finn ES, Scheinost D, et al. Using connectome-based predictive modeling to predict individual behavior from brain connectivity. Nat Protoc 2017;12(3):506-518.
Drysdale AT, Grosenick L, Downar J, et al. Resting-state connectivity biomarkers define neurophysiological subtypes of depression. Nat Med 2017;23(1):28-38.
Gowin JL, Manza P, Ramchandani VA, Volkow ND. Neuropsychosocial markers of binge drinking in young adults. Mol Psychiatry 2020;26(9):4931-4943.
Cheng K, Wang N, Li M. Interpretability of deep learning: A survey. In: Meng H, Lei T, Li M, Li K, Xiong N, Wang L, editors. Advances in natural computation, fuzzy systems and knowledge discovery. Cham: Springer International Publishing; 2021. p 475-486.
Khan AG, Hofer SB. Contextual signals in visual cortex. Curr Opin Neurobiol 2018;52:131-138.
Cavanna F, Vilas MG, Palmucci M, Tagliazucchi E. Dynamic functional connectivity and brain metastability during altered states of consciousness. Neuroimage 2018;180:383-395.
Zhou J, Cui G, Zhang Z, Yang C, Liu Z, Sun M. Graph neural networks: A review of methods and applications. AI Open 2018;1:57-81.
Barch DM, Burgess GC, Harms MP, et al. Function in the human connectome: Task-fMRI and individual differences in behavior. Neuroimage 2013;80:169-189.
Shen X, Tokoglu F, Papademetris X, Constable RT. Groupwise whole-brain parcellation from resting-state fMRI data for network node identification. Neuroimage 2013;82:403-415.
Rolls ET, Huang C-C, Lin C-P, Feng J, Joliot M. Automated anatomical labelling atlas 3. Neuroimage 2020;206:116189.
Rubinov M, Sporns O. Complex network measures of brain connectivity: Uses and interpretations. Neuroimage 2010;52(3):1059-1069.
Whelan R, Watts R, Orr CA, et al. Neuropsychosocial profiles of current and future adolescent alcohol misusers. Nature 2014;512(7513):185-189.
Crane NA, Gorka SM, Weafer J, Langenecker SA, de Wit H, Phan KL. Preliminary evidence for disrupted nucleus accumbens reactivity and connectivity to reward in binge drinkers. Alcohol Alcohol 2017;52(6):647-654.
Tarrant M, Smith J, Ball S, Winlove C, Gul S, Charles N. Alcohol consumption among university students in the night-time economy in the UK: A three-wave longitudinal study. Drug Alcohol Depend 2019;204:107522.
Lannoy S, Gilles F, Benzerouk F, et al. Disentangling the role of social cognition processes at early steps of alcohol abuse: The influence of affective theory of mind. Addict Behav 2020;102:106187.
Carbia C, Cadaveira F, Caamaño-Isorna F, Rodríguez-Holguín S, Corral M. Binge drinking during adolescence and young adulthood is associated with deficits in verbal episodic memory. PLoS One 2017;12(2):e0171393.
Hassing LB. Light alcohol consumption does not protect cognitive function: A longitudinal prospective study. Front Aging Neurosci 2018;10:81.
Katherine W, Carmen R. Language impairments in children with fetal alcohol spectrum disorders. J Popul Ther Clin Pharmacol 2011;18(2):364-376.
Aaron S. Americans and text messaging. In: Pew Internet & American Life Project, Vol 24. Washington, DC: Pew Research Center; 2011.
Buckner RL, Andrews-Hanna JR, Schacter DL. The Brain's default network. Ann N Y Acad Sci 2008;1124(1):1-38.
López-Caneda E, Cadaveira F, Correas A, Crego A, Maestú F, Rodríguez HS. The brain of binge drinkers at rest: Alterations in theta and Beta oscillations in first-year college students with a binge drinking pattern. Front Behav Neurosci 2017;11:168.
Fritz M, Klawonn AM, Zahr NM. Neuroimaging in alcohol use disorder: From mouse to man. J Neurosci Res 2019;100:1140-1158. https://doi.org/10.1002/jnr.24423.
Li G, Du S, Niu J, et al. Identification of binge drinkers via convolutional neural network and support vector machine. 2021 IEEE international conference on mechatronics and automation (ICMA). Takamatsu, Japan: Kagawa University; 2021. p 715-720.
Liu X, Lauer KK, Ward BD, et al. Fine-grained parcellation of brain connectivity improves differentiation of states of consciousness during graded propofol sedation. Brain Connect 2017;7(6):373-381.
Feilong M, Guntupalli JS, Haxby JV. The neural basis of intelligence in fine-grained cortical topographies. Elife 2021;10:e64058.
Liu R, Liu B-X, Ma M, et al. Aberrant prefrontal-parietal-cerebellar circuits in alcohol dependence. Neuropsychiatr Dis Treat 2018;14:3143-3150.
Bensmann W, Zink N, Werner A, Beste C, Stock A-K. Acute alcohol effects on response inhibition depend on response automatization, but not on GABA or glutamate levels in the ACC and striatum. J Clin Med 2020;9(2):481.
Weafer J, Gorka SM, Dzemidzic M, Kareken DA, Phan KL, de Wit H. Neural correlates of inhibitory control are associated with stimulant-like effects of alcohol. Neuropsychopharmacology 2021;46(8):1442-1450.
Zhornitsky S, Zhang S, Ide JS, et al. Alcohol expectancy and cerebral responses to Cue-elicited craving in adult nondependent drinkers. Biol Psychiatry Cogn Neurosci Neuroimag 2019;4(5):493-504.
Wang J, Fan Y, Dong Y, et al. Combining gray matter volume in the cuneus and the cuneus-prefrontal connectivity may predict early relapse in abstinent alcohol-dependent patients. PLoS One 2018;13(5):e0196860.
Roos A, Fouche J-P, Ipser JC, et al. Structural and functional brain network alterations in prenatal alcohol exposed neonates. Brain Imaging Behav 2021;15(2):689-699.
Seo D, Jia Z, Lacadie CM, Tsou KA, Bergquist K, Sinha R. Sex differences in neural responses to stress and alcohol context cues. Hum Brain Mapp 2011;32(11):1998-2013.
Fryer SL, Tapert SF, Mattson SN, Paulus MP, Spadoni AD, Riley EP. Prenatal alcohol exposure affects frontal-striatal BOLD response during inhibitory control. Alcohol Clin Exp Res 2007;31(8):1415-1424.
Müller-Oehring EM, Jung Y-C, Pfefferbaum A, Sullivan EV, Schulte T. The resting brain of alcoholics. Cereb Cortex 2015;25(11):4155-4168.