Search

Backgrounds Panic disorder is a highly prevalent psychiatric disorder that substantially impairs quality of life and psychosocial function. Panic disorder arises from neurobiological substrates and developmental factors that distinguish it from other anxiety disorders. Differential diagnosis between panic disorder and other anxiety disorders has only been conducted in terms of a phenomenological spectrum. Methods Through a machine learning-based approach with heart rate variability (HRV) as input, we aimed to build algorithms that can differentiate panic disorder from other anxiety disorders. Five algorithms were used: random forest (RF), gradient boosting machine (GBM), support vector machine (SVM), artificial neural network (ANN), and regularized logistic regression (LR). 10-fold cross-validation with five repeats was used to build the final models. Results A total of 60 patients with panic disorder and 61 patients with other anxiety disorders (aged between 20 and 65 years) were recruited. The L1-regularized LR showed the best accuracy (0.784), followed by ANN (0.730), SVM (0.730), GBM (0.676), and finally RF (0.649). LR also had good performance in other measures, such as F1-score (0.790), specificity (0.737), sensitivity (0.833), and Matthews correlation coefficient (0.572). Limitations Cross-sectional design and limited sample size is limitations. Conclusion This study demonstrated that HRV can be used to differentiate panic disorder from other anxiety disorders. Future studies with larger sample sizes and longitudinal design are required to replicate the diagnostic utility of HRV in a machine learning approach.

Background Suicide is a serious social and public health problem. Social stigma and prejudice reduce the accessibility of mental health care services for high-risk groups, resulting in them not receiving interventions and committing suicide. A suicide prediction model is necessary to identify high-risk groups in the general population. Methods We used national medical check-up data from 2009 to 2015 in Korea. The latest medical check-up data for each subject was set as an index point. Analysis was undertaken for an overall follow-up period (index point to the final tracking period) as well as for a one-year follow-up period. The training set was cross-validated fivefold. The predictive model was trained using a random forest algorithm, and its performance was measured using a separate test set not included in the training. Results The analysis covered 372,813 individuals, with an average (SD) overall follow-up duration of 1.52 (1.52) years. When we predicted suicide during the overall follow-up period, the area under the receiver operating characteristic curve (AUC) was 0.849, sensitivity was 0.817, and specificity was 0.754. The performance of the predicted suicide risk model for one year from the index point was AUC 0.818, sensitivity 0.788, and specificity 0.657. Conclusions This is probably the first suicide predictive model using machine learning based on medical check-up data from the general population. It could be used to screen high-risk suicidal groups from the population through routine medical check-ups. Future studies may test preventive interventions such as exercise and alcohol in these high-risk groups.

Background Many studies have reported that selective serotonin reuptake inhibitors (SSRI) are associated with an increased risk of bleeding. Mirtazapine and bupropion, which commonly lack serotonin reuptake inhibition, have been recommended as alternatives for patients who are at risk for bleeding. However, the evidence for these recommendations is insufficient. Methods We conducted a systematic search, systematic review, and meta-analysis to investigate an evidence-based approach for the bleeding risks of mirtazapine and bupropion. From 1946 to May 2017, a total of 3981 studies were searched from PubMed, Embase, and the Cochrane Library. Among the studies, two independent reviewers selected studies per predefined eligibility criteria. Results A total of five meta-analyses were conducted. Patients taking mirtazapine were at a greater risk of gastrointestinal bleeding (OR = 1.17, 95% CI = 1.01–1.38) than those who did not take antidepressants. No differences were observed in the bleeding risk between mirtazapine and SSRI or between bupropion and SSRI. Limitations The number of studies included in the meta-analysis was small. Conclusion Our results suggest that it is premature to recommend mirtazapine and bupropion for patients who have a bleeding risk. More studies with larger sample sizes and longitudinal follow-ups are warranted.

Background: Mental disorders play an important role both in clinical courses and recidivism among juvenile offenders. In view of the public health, prevalence of the mental disorders among the population is most fundamental element for planning strategy and establishing budget. Methods: A systematic search was conducted in the Embase, PubMed, PsycInfo, Web of Science, and Google scholar for studies reporting the prevalence rate of mental disorders in May 2018. Systematic review and meta-analysis was conducted to identify prevalence of mental disorders. Findings: A total 51,209 (32,616 boys and 8,393 girls) juvenile offenders from 90 studies in the 27 countries were included. The mean prevalence rate of any mental disorder was 0.791 (95% CI 0.786-0.796). The mean number of mental disorders per person was 2.1 in boys and 3.0 in girls. Boys (0.669 (95% CI 0.663-0.674)) had a higher prevalence rate than girls (0.530 (95% CI 0.515-0.546)). Depressive disorder (0.343 (95% CI 0.325-0.361)) vs 0.131 (95% CI 0.126-0.137)), bipolar disorder (0.374 (95% CI 0.349-0.399)) vs 0.163 (95% CI 0.157-0.170)), and posttraumatic stress disorder (0.243 (95% CI 0.225-0.261)) vs 0.083 (95% CI 0.781-0.882)), were more common in girls than in boys. The prevalence of psychotic disorder 0.060 (95% CI 0.056-0.064) was 30-fold higher than that in the general population. Interpretation: Mental disorders are substantially high among juvenile offenders. Active evaluation and intervention for the mental health are warranted particularly in offending girls. Funding: None. Declaration of Interest: We declare no competing interests.

Background In contrast to the previous studies reporting that most suicides occur among people with mental disorders, recent studies have reported various rates of mental disorders in suicide in different geographical regions. We aimed to comprehensively investigate the factors influencing the variation in the prevalence of mental disorders reported among suicide victims. Method The authors searched Embase, Medline, Web of Science, and the Cochrane Library to identify psychological autopsy studies reporting the prevalence of any mental disorders among suicide victims. A meta-regression analysis was conducted to identify the potential effects of geographical regions, the year of publication, measurements of personality disorder, measurements of comorbidity, and the ratio of females on the prevalence of mental disorders in addition to examining the heterogeneity across studies. Results From 4475 potentially relevant studies, 48 studies met eligibility criteria, with 6626 suicide victims. The studies from East Asia had a significantly lower mean prevalence (69.6% [95% CI=56.8 to 80.0]) than those in North America (88.2% [95% CI=79.7–93.5]) and South Asia (90.4% [95% CI=71.8–97.2]). The prevalence of any mental disorder decreased according to the year of publication (coefficients=−0.0715, p Limitations Substantial heterogeneities were identified within all subgroup analyses. Conclusions The prevalence of mental disorders among suicide cases seemed relatively low in the East Asia region, and recently published studies tended to report a lower prevalence of mental disorders. The link between the risk factors and suicide in the absence of a mental disorder should be examined in different geographical and sociocultural contexts.

Because depression has high prevalence and cause enduring disability, it is important to predict onset of depression among community dwelling adults. In this study, we aimed to build a machine learning-based predictive model for future onset of depression. We used nationwide survey data to construct training and hold-out test set. The class imbalance was dealt with the Synthetic Minority Over-sampling Technique. A tree-based ensemble method, random forest, was used to build a predictive model. Depression was defined by 9 or more on the Center for Epidemiologic Studies – Depression Scale 11 items version. Hyperparameters were tuned throughout the 10-fold cross-validation. A total of 6,588 (6,067 of non-depression and 521 of depression) participants were included in the study. The area under receiver operating characteristics curve was 0.870. The overall accuracy, sensitivity, and specificity were 0.862, 0.730, and 0.866, respectively. Satisfactions for leisure, familial relationship, general, social relationship, and familial income had importance in building predictive model for the onset of future depression. Our study demonstrated that predicting future onset of depression by using survey data could be possible. This predictive model is expected to be used for early identification of individuals at risk for depression and secure time to intervention.

The neurobiological causes underlying suicidal behaviors in major depressive disorder (MDD) have not been identified. This study was performed to investigate the differences in brain cortical thickness, surface area, and volume between suicide attempters and non-attempters with MDD. We performed magnetic resonance imaging (MRI) in 38 MDD patients (18-65 years old; 18 male, 20 female) with and without a history of suicide attempts. FreeSurfer software was used to compare the cortical thickness, surface area, and volume of 19 suicide attempters with MDD and 19 suicide non-attempters with MDD, while controlling for age, sex, mean area (or volume), and severity of depression. Compared with suicide non-attempters, suicide attempters with MDD exhibited a larger surface area in the left postcentral area and left lateral occipital area and a larger cortical volume in the left postcentral area and left lateral orbitofrontal area. Suicide attempters exhibited a smaller surface area in the left superior frontal area than suicide non-attempters. The present findings provide evidence for neuroanatomical risk factors of suicide in MDD. Further research to replicate these results and determine the mechanisms underlying these findings is needed.

The purpose of the present study was to examine the relationship between age and dopamine D(2) receptor availability in striatal subdivisions of young and middle-aged healthy subjects using high-resolution positron emission tomography (PET) with [(11)C]raclopride to better characterize the nature of age-related decrements in striatal D(2) receptor availability. Twenty-four healthy volunteers completed 3-Tesla magnetic resonance imaging and high-resolution [(11)C]raclopride PET scans. The analyses using linear and exponential models revealed that age had a significant negative correlation with D(2) receptor availability in the post-commissural putamen (postPU) and that D(2) receptor binding in the postPU decreased significantly more with age than in the ventral striatum, suggesting subregional differences in age-related changes in D(2) receptor binding. The postPU, which belongs to the sensorimotor striatum, may be particularly vulnerable to the effects of age in young and middle-aged subjects.