Search

Background: No previous study has comprehensively reviewed the application of machine learning algorithms in mood disorders populations. Herein, we qualitatively and quantitatively evaluate previous studies of machine learning-devised models that predict therapeutic outcomes in mood disorders populations., Methods: We searched Ovid MEDLINE/PubMed from inception to February 8, 2018 for relevant studies that included adults with bipolar or unipolar depression; assessed therapeutic outcomes with a pharmacological, neuromodulatory, or manual-based psychotherapeutic intervention for depression; applied a machine learning algorithm; and reported predictors of therapeutic response. A random-effects meta-analysis of proportions and meta-regression analyses were conducted., Results: We identified 639 records: 75 full-text publications were assessed for eligibility; 26 studies (n=17,499) and 20 studies (n=6325) were included in qualitative and quantitative review, respectively. Classification algorithms were able to predict therapeutic outcomes with an overall accuracy of 0.82 (95% confidence interval [CI] of [0.77, 0.87]). Pooled estimates of classification accuracy were significantly greater (p < 0.01) in models informed by multiple data types (e.g., composite of phenomenological patient features and neuroimaging or peripheral gene expression data; pooled proportion [95% CI] = 0.93[0.86, 0.97]) when compared to models with lower-dimension data types (pooledproportion=0.68[0.62,0.74]to0.85[0.81,0.88])., Limitations: Most studies were retrospective; differences in machine learning algorithms and their implementation (e.g., cross-validation, hyperparameter tuning); cannot infer importance of individual variables fed into learning algorithm., Conclusions: Machine learning algorithms provide a powerful conceptual and analytic framework capable of integrating multiple data types and sources. An integrative approach may more effectively model neurobiological components as functional modules of pathophysiology embedded within the complex, social dynamics that influence the phenomenology of mental disorders., (Copyright © 2018. Published by Elsevier B.V.)

Background: Despite the widespread use of antidepressants in clinical practice, the current trial-and-error approach to medication selection contributes to treatment failure and underscores the need to identify reliable predictors of antidepressant response. Since changes in measures of cognition have been reported to occur early in treatment and prior to improvements in overall mood symptoms, the present review aims to determine whether early changes in measures of cognition can predict response in individuals with MDD., Methods: A systematic review of studies evaluating early cognitive change as a predictor of later treatment response in MDD was conducted using PubMed/Medline, Embase and PsychINFO., Results: A total of seven articles were identified. The available evidence suggests the early changes in cognition may predict treatment response in individuals with MDD. This was shown across antidepressant classes (i.e., SSRIs, SNRIs, NRIs, melatonergic antidepressants) and forms of therapy (i.e., pharmacotherapy, rTMS). The results depict an emerging trend towards early changes in facial emotion recognition (i.e., a hot cognitive process) as a predictor of treatment outcome., Limitations: Our qualitative analysis reflects a very limited number of studies. Moreover, there was significant heterogeneity in the evaluation of cognition across studies. Future research should aim to parse out this heterogeneity by evaluating the relative predictive value of different measures of cognition., Conclusion: The identification of reliable early treatment predictors of antidepressant response would be clinically significant, enabling clinicians to more accurately evaluate the efficacy of selected treatment avenues., (Copyright © 2018 Elsevier B.V. All rights reserved.)

Objectives: Extant evidence indicates that ketamine exerts rapid antidepressant effects in treatment-resistant depressive (TRD) symptoms as a part of major depressive disorder (MDD) and bipolar disorder (BD). The identification of depressed sub-populations that are more likely to benefit from ketamine treatment remains a priority. In keeping with this view, the present narrative review aims to identify the pretreatment predictors of response to ketamine in TRD as part of MDD and BD., Method: Electronic search engines PubMed/MEDLINE, ClinicalTrials.gov, and Scopus were searched for relevant articles from inception to January 2018. The search term ketamine was cross-referenced with the terms depression, major depressive disorder, bipolar disorder, predictors , and response and/or remission ., Results: Multiple baseline pretreatment predictors of response were identified, including clinical (i.e., Body Mass Index (BMI), history of suicide, family history of alcohol use disorder), peripheral biochemistry (i.e., adiponectin levels, vitamin B12 levels), polysomnography (abnormalities in delta sleep ratio), neurochemistry (i.e., glutamine/glutamate ratio), neuroimaging (i.e., anterior cingulate cortex activity), genetic variation (i.e., Val66Met BDNF allele), and cognitive functioning (i.e., processing speed). High BMI and a positive family history of alcohol use disorder were the most replicated predictors., Conclusions: A pheno-biotype of depression more, or less likely, to benefit with ketamine treatment is far from complete. Notwithstanding, metabolic-inflammatory alterations are emerging as possible pretreatment response predictors of depressive symptom improvement, most notably being cognitive impairment. Sophisticated data-driven computational methods that are iterative and agnostic are more likely to provide actionable baseline pretreatment predictive information.