Random Forest Classification of Alcohol Use Disorder Using fMRI Functional Connectivity, Neuropsychological Functioning, and Impulsivity Measures

Abstract

Individuals with alcohol use disorder (AUD) are known to manifest a variety of neurocognitive impairments that can be attributed to alterations in specific brain networks. The current study aims to identify specific features of brain connectivity, neuropsychological performance, and impulsivity traits that can classify adult males with AUD (n = 30) from healthy controls (CTL, n = 30) using the Random Forest (RF) classification method. The predictor variables were: (i) fMRI-based within-network functional connectivity (FC) of the Default Mode Network (DMN), (ii) neuropsychological scores from the Tower of London Test (TOLT), and the Visual Span Test (VST), and (iii) impulsivity factors from the Barratt Impulsiveness Scale (BIS). The RF model, with a classification accuracy of 76.67%, identified fourteen DMN connections, two neuropsychological variables (memory span and total correct scores of the forward condition of the VST), and all impulsivity factors as significantly important for classifying participants into either the AUD or CTL group. Specifically, the AUD group manifested hyperconnectivity across the bilateral anterior cingulate cortex and the prefrontal cortex as well as between the bilateral posterior cingulate cortex and the left inferior parietal lobule, while showing hypoconnectivity in long-range anterior–posterior and interhemispheric long-range connections. Individuals with AUD also showed poorer memory performance and increased impulsivity compared to CTL individuals. Furthermore, there were significant associations among FC, impulsivity, neuropsychological performance, and AUD status. These results confirm the previous findings that alterations in specific brain networks coupled with poor neuropsychological functioning and heightened impulsivity may characterize individuals with AUD, who can be efficiently identified using classification algorithms such as Random Forest.

Random Forest Classification of Alcohol Use Disorder Using EEG Source Functional Connectivity, Neuropsychological Functioning, and Impulsivity Measures

“…Since beta oscillations are associated with AUD [155] and other externalizing traits [156], associations among beta band FC, impulsivity and hyperconnectivity in AUD are in line with the view that AUD is primarily a disinhibitory disorder [111] and may be related to neuronal hyperexcitability [108]. Interestingly, similar to the current findings, impulsivity factors ranked as the topmost predictors in our previous study on fMRI FC with the same set of participants [54]. Earlier studies have also drawn etiological connections among AUD, externalizing traits such as impulsivity, and neural disinhibition in the form of electrophysiological features in AUD individuals (e.g., low P3 amplitude and delta and theta oscillations underlying P3 during cognitive processing, and increased resting state beta power) [108,[118][119][120].…”

“…The DMN regions analyzed in the study are posterior cingulate cortex (PCC), anterior cingulate cortex (ACC), inferior parietal cortex (IPL), prefrontal cortices (PFC), lateral temporal cortex (LTC), and hippocampal formation (HCF) ( Table 2 and Figure 1), in line with the fc-MRI and fc-EEG studies [58,[78][79][80]. More information about the DMN seed selection is described in our recent publication with fc-MRI in the same sample as the current study [54]. Each seed region contained the voxels within 10 mm radius from the peak/centroid point of the region.…”

“…Recently, in another fMRI FC study using random forest (RF), Zhu et al [29] reported that executive control networks showed the strongest prediction accuracy among the within-network features, while the connectivity between DMN and reward network as well as between executive control network and reward network contributed more to the predictive accuracy in classifying AUD from controls. Recently, using the same groups of individuals as the current study, we have also reported aberrant fMRI based FC across the DMN regions in AUD individuals [54]. In sum, these fMRI based FC studies in resting state revealed aberrations in AUD in DMN and other resting state networks.…”

“…Interestingly, in our previous study with the same groups of subjects, we reported that that lower volumes in prefrontal cortex and left hippocampus observed in AUD group were associated with poorer visuospatial memory performance [16]. Furthermore, the hyperconnectivity across parahippocampal hub in the current study as well as in our fMRI FC study with the same sample of individuals [54] may also be related to the deficits in visual memory in AUD subjects, possibly representing a compensatory mechanism during the memory performance. It is likely that executive deficits may have improved, as reflected in normal TOLT performance, possibly due to prolonged abstinence from drinking in majority of the AUD individuals.…”

“…The demographic and clinical characteristics of the sample are presented in Table 1, and a detailed sample description is available in Pandey et al [16] and Kamarajan et al [54]. Briefly, the sample comprised thirty male participants with AUD [mean age (standard deviation, SD) = 41.42 (7.31) years] and thirty unaffected male controls [mean age (SD) = 27.44 (4.74) years].…”