Timothy Durazzo, Postdoctoral Faculty Sponsor
Regional cortical brain volumes at treatment entry relates to post treatment WHO risk drinking levels in those with alcohol use disorder.
Drug and alcohol dependence
2024; 255: 111082
Abstinence following treatment for alcohol use disorder (AUD) is associated with significant improvements in psychiatric and physical health, however, recent studies suggest resumption of low risk levels of alcohol use can also be beneficial. The present study assessed whether post-treatment levels of alcohol use were associated with cortical brain volumedifferences at treatment entry.Individuals seeking treatment for AUD (n=75) and light/non-drinking controls (LN, n=51) underwent 1.5T magnetic resonance imaging. The volumes of 34 bilateral cortical regions of interest (ROIs) were quantitated via FreeSurfer. Individuals with AUD were classified according to post-treatment alcohol consumption using the WHO risk drinking levels (abstainers: AB; low risk: RL; or higher risk: RH). Regional volumes for AB, RL and RH, at treatment entry, were compared to LN.Relative to LN, AB demonstrated smaller volumes in 18/68 (26%), RL in 24/68 (35%) and RH in 34/68 (50%) ROIs with the largest magnitude volume differences observed between RH and LN. RH and RL reported a higher frequency of depressive disorders than AB. Among RH and RL, level of depressive and anxiety symptomatology were associated with daily number of drinks consumed after treatment.Volumetric differences, at treatment entry, in brain regions implicated in executive function and salience networks corresponded with post-treatment alcohol consumption levels suggesting that pre-existing differences in neural integrity may contribute to treatment outcomes. Depressive and anxiety symptomatology was also associated with brain morphometrics and alcohol use patterns, highlighting the importance of effectively targeting these conditions during AUD treatment.
View details for DOI 10.1016/j.drugalcdep.2024.111082
View details for PubMedID 38219355
Using neuroimaging to predict relapse in stimulant dependence: A comparison of linear and machine learning models.
2019; 21: 101676
OBJECTIVE: Relapse rates are consistently high for stimulant user disorders. In order to obtain prognostic information about individuals in treatment, machine learning models have been applied to neuroimaging and clinical data. Yet few efforts have been made to test these models in independent samples or show that they can outperform linear models. In this exploratory study, we examine whether machine learning models relative to linear models provide greater predictive accuracy and less overfitting.METHOD: This longitudinal study included 63 methamphetamine-dependent (training sample) and 29 cocaine-dependent (test sample) individuals who completed an MRI scan during residential treatment. Linear and machine learning models predicting relapse at a one-year follow up that were previously developed in the methamphetamine-dependent sample using neuroimaging and clinical variables were applied to the cocaine-dependent sample. Receiver operating characteristic analysis was used to assess performance using area under the curve (AUC) as the primary outcome.RESULTS: Twelve individuals in the cocaine-dependent sample remained abstinent, and 17 relapsed. The linear models produced more accurate prediction in the training sample than the machine learning models but showed reduced performance in the testing sample, with AUC decreasing by 0.18. The machine learning models produced similar predictive performance in the training and test samples, with AUC changing by 0.03. In the test sample, neither the linear nor the machine learning model predicted relapse at rates above chance.CONCLUSIONS: Although machine learning algorithms may have advantages, in this study neither model's performance was sufficient to be clinically useful. In order to improve predictive models, stronger predictor variables and larger samples are needed.
View details for PubMedID 30665102
- Using neuroimaging to predict relapse in stimulant dependence: A comparison of linear and machine learning models NEUROIMAGE-CLINICAL 2019; 21