Neuroanatomical features reveal accelerated brain age in alcohol use disorder
Kamarajan, C.; Ardekani, B. A.; Pandey, A. K.; Meyers, J. L.; Chorlian, D. B.; Kinreich, S.; Pandey, G.; Saenz de Viteri, S.; Richard, C.; Bingly, A.; Kuang, W.; Porjesz, B.
Show abstract
Background/ObjectivesBrain age is a novel measure to characterize the integrity of neurocognitive functioning and brain health in various psychiatric and neurological disorders. Although there is a literature suggesting premature aging of the brain in individuals with alcohol use disorder (AUD), studies directly examining brain age are rare. Therefore, the current study was designed to estimate brain age in AUD individuals using brain morphological features, such as cortical thickness and brain volume. MethodsThe sample included a group of 30 adult males with a history of AUD but maintaining abstinence and a group of 30 male controls without any history of AUD. Brain age was computed using an XGBoost regression model with 187 brain morphological features of cortical thickness and brain volume as predictors. An exploratory correlational analysis between brain age measures and features of neuropsychological performance, impulsivity, and alcohol consumption was also performed. ResultsFindings revealed that AUD individuals showed an increase of 1.70 years in their brain age relative to their chronological age. Further, in the AUD group, higher brain age was significantly associated with poor executive functioning, while a larger gap between brain age and actual age was associated with lower non-planning impulsivity and later age of onset for regular drinking in those with AUD. ConclusionsAUD individuals manifested accelerated brain aging, possibly representing compromised brain health. Brain age measures were found to be associated with some of the measures of neurocognition, impulsivity, and alcohol consumption. These findings may have important implications for the early identification, prevention, and treatment of AUD. However, future studies with larger sample sizes are needed to confirm these preliminary findings.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Proteomic associations with cognitive variability as measured by the Wisconsin Card Sorting Test in a healthy Thai population: A machine learning approach 95%
- Behavioural cue reactivity to alcohol and non-alcohol-related stimuli among individuals with severe alcohol use disorder: an fMRI study with a visual task 94%
- Longer chronic cannabis use in humans is associated with impaired implicit motor learning and supranormal resting state cortical activity 93%
Similar papers in this journal
- Reliability and Validity of Cognitive Workload in Older Adults 93%
- Lower nerve growth factor levels in major depression and suicidal behaviors: effects of adverse childhood experiences and recurrence of illness 92%
- Abnormal Functional Network Topology and its Dynamics during Sustained Attention Processing Significantly Implicate Post-TBI Attention Deficits in Children 92%
Similar papers in this journal
- Gender score development in a retrospective approach in the Berlin Aging Study II 93%
- Menarche, pubertal timing and the brain: female-specific patterns of brain maturation beyond age-related development 91%
- Sex differences in contextual pattern separation, activation of new neurons, and functional connectivity within the limbic system 90%
Similar papers in this journal
- mtDNA copy number, mtDNA total somatic deletions, Complex I activity, synapse number and synaptic mitochondria number are altered in schizophrenia and bipolar disorder 93%
- Maternal immune activation and adolescent alcohol exposure increases alcohol drinking and disrupts cortical-striatal-hippocampal oscillations in adult offspring 92%
- MRI signature of brain age underlying post-traumatic stress disorder in World Trade Center responders 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.