Development of Prediction Model for Suicide Attempts Using the Korean Youth Health Behavior Web-Based Survey in Korean Middle and High School Students / 신경정신의학

Objectives@#Assessing the risks of youth suicide in educational and clinical settings is crucial.Therefore, this study developed a machine learning model to predict suicide attempts using the Korean Youth Risk Behavior Web-based Survey (KYRBWS). @*Methods@#KYRBWS is conducted annually on Korean middle and high school students to assess their health-related behaviors. The KYRBWS data for 2021, which showed 1206 adolescents reporting suicide attempts out of 54848, was split into the training (n=43878) and test (n=10970) datasets. Thirty-nine features were selected from the KYRBWS questionnaire. The balanced accuracy of the model was employed as a metric to select the best model. Independent validations were conducted with the test dataset of 2021 KYRBWS (n=10970) and the external dataset of 2020 KYRBWS (n=54948). The clinical implication of the prediction by the selected model was measured for sensitivity, specificity, true prediction rate (TPR), and false prediction rate (FPR). @*Results@#Balanced bag of histogram gradient boosting model has shown the best performance (balanced accuracy=0.803). This model shows 76.23% sensitivity, 83.08% specificity, 10.03% TPR, and 99.30% FPR for the test dataset as well as 77.25% sensitivity, 84.62% specificity, 9.31% TPR, and 99.45% FPR for the external dataset, respectively. @*Conclusion@#These results suggest that a specific machine learning model can predict suicide attempts among adolescents with high accuracy.

Relationship Between the Loudness Dependence of the Auditory Evoked Potential and the Severity of Suicidal Ideation in Patients with Major Depressive Disorder

Objective@#The loudness dependence of the auditory evoked potential (LDAEP) is a reliable indicator that is inversely related to central serotonergic activity, and recent studies have suggested an association between LDAEP and suicidal ideation. This study investigated differences in LDAEP between patients with major depressive disorder and high suicidality and those with major depressive disorder and low suicidality compared to healthy controls. @*Methods@#This study included 67 participants: 23 patients with major depressive disorder with high suicidality (9 males, mean age 29.3 ± 15.7 years, total score of SSI-BECK ≥ 15), 22 patients with major depressive disorder with low suicidality (9 males, mean age 42.2 ± 14.4 years, total score of SSI-BECK ≤ 14), and 22 healthy controls (11 males, mean age 31.6 ± 8.7 years). Participants completed the following assessments: Patient Health Questionnaire-9, Beck Depression Inventory-II, Beck Scale for Suicidal ideation, State Anxiety Scale of the State-Trait Anxiety Inventory, Beck Anxiety Inventory, and LDAEP (measured at electrode Cz). @*Results@#There were no sex-related differences among groups (p = 0.821). The high-suicidality group exhibited significantly higher LDAEP compared to the low-suicidality group (0.82 ± 0.79 vs. 0.26 ± 0.36, p = 0.014). No significant differences were found between the control and high-suicidality (p = 0.281) or the control and low-suicidality groups (p = 0.236). @*Conclusion@#LDAEP was applied to demonstrate the association between serotonergic activity and suicidal ideation and suicide risk in major depression and may be a candidate of biological marker for preventing suicide in this study.

Relationship Between the Loudness Dependence of the Auditory Evoked Potential and the Severity of Suicidal Ideation in Patients with Major Depressive Disorder

Objective@#The loudness dependence of the auditory evoked potential (LDAEP) is a reliable indicator that is inversely related to central serotonergic activity, and recent studies have suggested an association between LDAEP and suicidal ideation. This study investigated differences in LDAEP between patients with major depressive disorder and high suicidality and those with major depressive disorder and low suicidality compared to healthy controls. @*Methods@#This study included 67 participants: 23 patients with major depressive disorder with high suicidality (9 males, mean age 29.3 ± 15.7 years, total score of SSI-BECK ≥ 15), 22 patients with major depressive disorder with low suicidality (9 males, mean age 42.2 ± 14.4 years, total score of SSI-BECK ≤ 14), and 22 healthy controls (11 males, mean age 31.6 ± 8.7 years). Participants completed the following assessments: Patient Health Questionnaire-9, Beck Depression Inventory-II, Beck Scale for Suicidal ideation, State Anxiety Scale of the State-Trait Anxiety Inventory, Beck Anxiety Inventory, and LDAEP (measured at electrode Cz). @*Results@#There were no sex-related differences among groups (p = 0.821). The high-suicidality group exhibited significantly higher LDAEP compared to the low-suicidality group (0.82 ± 0.79 vs. 0.26 ± 0.36, p = 0.014). No significant differences were found between the control and high-suicidality (p = 0.281) or the control and low-suicidality groups (p = 0.236). @*Conclusion@#LDAEP was applied to demonstrate the association between serotonergic activity and suicidal ideation and suicide risk in major depression and may be a candidate of biological marker for preventing suicide in this study.

Application of Text-Classification Based Machine Learningin Predicting Psychiatric Diagnosis

Objectives@#ZZThe aim was to find effective vectorization and classification models to predict a psychiatric diagnosis from text-basedmedical records. @*Methods@#ZZElectronic medical records (n = 494) of present illness were collected retrospectively in inpatient admission notes withthree diagnoses of major depressive disorder, type 1 bipolar disorder, and schizophrenia. Data were split into 400 training data and 94 independentvalidation data. Data were vectorized by two different models such as term frequency-inverse document frequency (TF-IDF)and Doc2vec. Machine learning models for classification including stochastic gradient descent, logistic regression, support vectorclassification, and deep learning (DL) were applied to predict three psychiatric diagnoses. Five-fold cross-validation was used to find aneffective model. Metrics such as accuracy, precision, recall, and F1-score were measured for comparison between the models. @*Results@#ZZFive-fold cross-validation in training data showed DL model with Doc2vec was the most effective model to predict the diagnosis(accuracy = 0.87, F1-score = 0.87). However, these metrics have been reduced in independent test data set with final workingDL models (accuracy = 0.79, F1-score = 0.79), while the model of logistic regression and support vector machine with Doc2vec showedslightly better performance (accuracy = 0.80, F1-score = 0.80) than the DL models with Doc2vec and others with TF-IDF. @*Conclusions@#ZZThe current results suggest that the vectorization may have more impact on the performance of classification thanthe machine learning model. However, data set had a number of limitations including small sample size, imbalance among the category,and its generalizability. With this regard, the need for research with multi-sites and large samples is suggested to improve the machinelearning models.

Application of Text-Classification Based Machine Learningin Predicting Psychiatric Diagnosis

Objectives@#ZZThe aim was to find effective vectorization and classification models to predict a psychiatric diagnosis from text-basedmedical records. @*Methods@#ZZElectronic medical records (n = 494) of present illness were collected retrospectively in inpatient admission notes withthree diagnoses of major depressive disorder, type 1 bipolar disorder, and schizophrenia. Data were split into 400 training data and 94 independentvalidation data. Data were vectorized by two different models such as term frequency-inverse document frequency (TF-IDF)and Doc2vec. Machine learning models for classification including stochastic gradient descent, logistic regression, support vectorclassification, and deep learning (DL) were applied to predict three psychiatric diagnoses. Five-fold cross-validation was used to find aneffective model. Metrics such as accuracy, precision, recall, and F1-score were measured for comparison between the models. @*Results@#ZZFive-fold cross-validation in training data showed DL model with Doc2vec was the most effective model to predict the diagnosis(accuracy = 0.87, F1-score = 0.87). However, these metrics have been reduced in independent test data set with final workingDL models (accuracy = 0.79, F1-score = 0.79), while the model of logistic regression and support vector machine with Doc2vec showedslightly better performance (accuracy = 0.80, F1-score = 0.80) than the DL models with Doc2vec and others with TF-IDF. @*Conclusions@#ZZThe current results suggest that the vectorization may have more impact on the performance of classification thanthe machine learning model. However, data set had a number of limitations including small sample size, imbalance among the category,and its generalizability. With this regard, the need for research with multi-sites and large samples is suggested to improve the machinelearning models.

Now and Future of Data Sharing : Brain Magnetic Resonance Imaging Repositories / 생물치료정신의학

Over the past decade, practice of sharing brain magnetic resonance imaging (MRI) data is increasing given significance of reproducibility and transparency in human neuroscience. Larger multimodal brain MRI databases are needed for more robust research findings considering potential possibilities of large variability in human neuroscience. There are currently more than tens of thousands of shared brain MRI datasets across multiple conditions and hundreds of neuroimaging studies using multimodality through shared brain MRI data repositories. This article critically reviews aims, procedures, and current state of brain MRI data sharing. This review focuses on projects and research findings using structural and functional MRI open databases and is further divided into T1- and diffusion-weighted images for structural MRI as well as resting-state and task-based functional MRI. The challenges and directions are finally discussed. Advances in brain MRI data sharing will lead to more rapid progression in human neuroscience by fostering effective longitudinal, multi-site, multimodal neuroimaging research.

Association Analysis between Chromogranin B Genetic Variations and Smooth Pursuit Eye Movement Abnormality in Korean Patients with Schizophrenia

OBJECTIVES: According to previous studies, the Chromogranin B (CHGB) gene could be an important candidate gene for schizophrenia which is located on chromosome 20p12.3. Some studies have linked the polymorphism in CHGB gene with the risk of schizophrenia. Meanwhile, smooth pursuit eye movement (SPEM) abnormality has been regarded as one of the most consistent endophenotype of schizophrenia. In this study, we investigated the association between the polymorphisms in CHGB gene and SPEM abnormality in Korean patients with schizophrenia. METHODS: We measured SPEM function in 24 Korean patients with schizophrenia (16 male, 8 female) and they were divided according to SPEM function into two groups, good and poor SPEM function groups. We also investigated genotypes of polymorphisms in CHGB gene in each group. A logistic regression analysis was performed to find the association between SPEM abnormality and the number of polymorphism. RESULTS: The natural logarithm value of signal/noise ratio (Ln S/N ratio) of good SPEM function group was 4.19 ± 0.19 and that of poor SPEM function group was 3.17 ± 0.65. In total, 15 single nucleotide polymorphisms of CHGB were identified and the genotypes were divided into C/C, C/R, and R/R. Statistical analysis revealed that two genetic variants (rs16991480, rs76791154) were associated with SPEM abnormality in schizophrenia (p = 0.004). CONCLUSIONS: Despite the limitations including a small number of samples and lack of functional study, our results suggest that genetic variants of CHGB may be associated with SPEM abnormality and provide useful preliminary information for further study.

Gender-Specific Associations between CHGB Genetic Variants and Schizophrenia in a Korean Population

PURPOSE: Schizophrenia is a devastating mental disorder and is known to be affected by genetic factors. The chromogranin B (CHGB), a member of the chromogranin gene family, has been proposed as a candidate gene associated with the risk of schizophrenia. The secretory pathway for peptide hormones and neuropeptides in the brain is regulated by chromogranin proteins. The aim of this study was to investigate the potential associations between genetic variants of CHGB and schizophrenia susceptibility. MATERIALS AND METHODS: In the current study, 15 single nucleotide polymorphisms of CHGB were genotyped in 310 schizophrenia patients and 604 healthy controls. RESULTS: Statistical analysis revealed that two genetic variants (non-synonymous rs910122; rs2821 in 3′-untranslated region) were associated with schizophrenia [minimum p=0.002; odds ratio (OR)=0.72], even after correction for multiple testing (p(corr)=0.02). Since schizophrenia is known to be differentially expressed between sexes, additional analysis for sex was performed. As a result, these two genetic variants (rs910122 and rs2821) and a haplotype (ht3) showed significant associations with schizophrenia in male subjects (p(corr)=0.02; OR=0.64), whereas the significance disappeared in female subjects (p>0.05). CONCLUSION: Although this study has limitations including a small number of samples and lack of functional study, our results suggest that genetic variants of CHGB may have sex-specific effects on the risk of schizophrenia and provide useful preliminary information for further study.

No Association between (AAT)n Repeat Polymorphisms in the Cannabinoid Receptor 1 Gene and Smooth Pursuit Eye Movement Abnormality in Korean Patients with Schizophrenia

OBJECTIVES: According to previous studies, the cannabinoid receptor 1 (CNR1) gene could be an important candidate gene for schizophrenia. Some studies have linked the (AAT)n trinucleotide repeat polymorphism in CNR1 gene with the risk of schizophrenia. Meanwhile, smooth pursuit eye movement (SPEM) has been regarded as one of the most consistent endophenotypes of schizophrenia. In this study, we investigated the association between the (AAT)n trinucleotide repeats in CNR1 gene and SPEM abnormality in Korean patients with schizophrenia. METHODS: We measured SPEM function in 167 Korean patients with schizophrenia (84 male, 83 female) and they were divided according to SPEM function into two groups, good and poor SPEM function groups. We also investigated allele frequencies of (AAT)n repeat polymorphisms on CNR1 gene in each group. A logistic regression analysis was performed to find the association between SPEM abnormality and the number of (AAT)n trinucleotide repeats. RESULTS: The natural logarithm value of signal/noise ratio (Ln S/N ratio) of the good SPEM function group was 4.34 ± 0.29 and that of the poor SPEM function group was 3.21 ± 0.70. In total, 7 types of trinucleotide repeats were identified, each containing 7, 10, 11, 12, 13, 14, and 15 repeats, respectively. In the patients with (AAT)₇ allele, the distributions of the good and poor SPEM function groups were 18 (11.1%) and 19 (11.0%) respectively. In the patients with (AAT)₁₀ allele, (AAT)₁₁ allele, (AAT)₁₂ allele, (AAT)₁₃ allele, (AAT)₁₄ allele and (AAT)₁₅ allele, the distributions of good and poor SPEM function groups were 13 (8.0%) and 12 (7.0%), 4 (2.5%) and 6 (3.5%), 31 (19.8%) and 35 (20.3%), 51 (31.5%) and 51 (29.7%), 36 (22.2%) and 45 (26.2%), 9 (5.6%) and 4 (2.3%) respectively. As the number of (AAT) n repeat increased, there was no aggravation of abnormality of SPEM function. CONCLUSIONS: There was no significant aggravation of SPEM abnormality along with the increase of number of (AAT)n trinucleotide repeats in the CNR1 gene in Korean patients with schizophrenia.

Influence of Depression on Working Memory Measured by Digit Backward Span in the Subjects with Mild Cognitive Impairment and Dementia

OBJECTIVES: We aimed to explore the influence of depression on working memory in patients with mild cognitive impairment (MCI) and dementia. METHODS: Clinical and neuropsychological data of 43 subjects with mild cognitive impairment (MCI) (n=17) and dementia (n=26) who had visited Department of Psychiatry at Soonchunhyang University Seoul Hospital, were collected. The subjects were divided into depressed (n=18) and non-depressed (n=25) groups based on the Korean version of Short Geriatric Depression Scale. Two-way analysis of variance test was conducted to evaluate the influence of diagnosis (MCI and dementia), the presence of depression and their interaction on working memory which was measured by digit forward and backward span test. RESULTS: Among the patients with MCI, test score of digit backward span test in depressed group was significantly lower than in non-depressed group. However, among the patients with dementia, there was no significant difference in digit backward span test between depressed and non-depressed groups. CONCLUSION: This study suggests that the depression could deteriorate working memory measured by digit backward span test in patients with MCI, relative to in patients with dementia and it also implicates the diagnostic assessment for depression has clinically importance in patients with MCI.

Association Analysis between (AAT)n Repeats in the Cannabinoid Receptor 1 Gene and Schizophrenia in a Korean Population

OBJECTIVES: Previous studies suggest that the cannabinoid receptor 1 (CNR1) gene could be an important candidate gene for schizophrenia. According to linkage studies, this gene is located on chromosome 6q14-q15, which is known to harbor the schizophrenia susceptibility locus (locus 5, SCZ5, OMIM 803175). The pharmacological agent delta-9-tetrahydrocannabinol (Delta-9-THC) seems to elicit the symptoms of schizophrenia. The association between CNR1 polymorphisms and schizophrenia is actively being investigated, and some studies have linked the AAT-trinucleotide repeats in CNR1 to the onset of schizophrenia. In this study, we have investigated the association between the AAT-trinucleotide repeats in CNR1 and schizophrenia by studying schizophrenia patients and healthy individuals from Korea. METHODS: DNA was extracted from the blood samples of 394 control subjects and 337 patients diagnosed with schizophrenia (as per the Diagnostic and Statistical Manual of Mental Disorders, fourth edition criteria). After polymerase chain reaction amplification, a logistic regression analysis, with age and gender as the covariates, was performed to study the variations in the AAT-repeat polymorphisms between the two groups. RESULTS: In total, 8 types of trinucleotide repeats were identified, each containing 7, 8, 10, 11, 12, 13, 14, and 15 repeats, respectively. (AAT)13 allele was most frequently observed, with a frequency of 33.6% and 31.6% in the patient and control groups, respectively. The frequency of the other repeat alleles in the patient group (in the decreasing order) was as follows : (AAT)13 33.6%, (AAT)14 21.6%, (AAT)12 18.5%, and (AAT)7 11.1%. The frequency of the repeat alleles in the control group (in the decreasing order) was as follows : (AAT)13 31.6%, (AAT)14 24.5%, (AAT)12 17.2%, and (AAT)7 11.6%. However, there were no significant differences in the AAT-repeat polymorphisms of the CNR1 gene between the patient group and the control group. CONCLUSIONS: Although our study revealed no significant association of the AAT-repeat polymorphism of the CNR1 gene with schizophrenia, it will serve as a good reference for future studies designed to examine the cannabinoid hypothesis of schizophrenia.

Association between Characteristics of Brain Magnetic Resonance Imaging and Atypical Antipsychotics Use in Dementia Patients

OBJECTIVES: We aimed to identify the neuroimaging marker for prediction of the use of atypical antipsychotics (AAP) in dementia patients. METHODS: From April 2010 to March 2013, 31 patients who were diagnosed as dementia at the psychiatric department of Soonchunhyang University Hospital, completed the brain magnetic resonance imaging scan and cognitive test for dementia. Ten patients were treated with AAP for the improvement of behavioral and psychological symptoms of dementia (BPSD) and the other 21patients were not. Using T1 weighted and Fluid Attenuated Inversion Recovery (FLAIR) images of brain, areas of white matter (WM), gray matter (GM), cerebrospinal fluid (CSF) and white matter hyperintensities (WMH) have been segmented and measured. Multivariate logistic regression models were applied for assessment of association between AAP use and the GM/WM ratio, the WMH/whole brain (GM + WM + CSF) ratio. RESULTS: There was a significant association between AAP use and the GM/WM ratio (odds ratio, OR = 1.18, 95% confidence interval, CI 1.01-1.38, p = 0.037), while there was no association between AAP use and the WMH/whole brain ratio (OR = 0.82, 95% CI 0.27-2.48, p = 0.73). CONCLUSIONS: The GM/WM ratio could be a biological marker for the prediction of AAP use and BPSD in patients with dementia. It was more likely to increase as dementia progress since atrophy of WM was more prominent than that of GM over aging.

Association Analysis of Neuregulin 3 Gene with Schizophrenia in a Korean Population

OBJECTIVES: Located on chromosome 10q22-q23, the human neuregulin 3 (NRG3) is suggested as a strong positional and functional candidate gene involved in the pathogenesis of schizophrenia. Several case-control studies examining the association between polymorphisms on NRG3 gene with schizophrenia and/or its traits (such as delusion) have been reported recently in cohorts of Han Chinese, Ashkenazi Jews, Australians, white Americans of Western European ancestry and Koreans. Thus, this study aimed to investigate the association of one SNP in exon 9 (rs2295933) of NRG3 gene with the risk of schizophrenia in a Korean population. METHODS: Using TaqMan assay, rs2295933 in the exon 9 of NRG3 was genotyped in 435 patients with schizophrenia as cases and 393 unrelated healthy individuals as controls. Differences in frequency distributions were analyzed using logistic regression models following various modes of genetic inheritance and controlling for age and sex as covariates. RESULTS: Subsequent analysis revealed that the frequency distribution of rs2295933 of NRG3 was not different between schizophrenia patients and healthy controls of Korean ethnicity. CONCLUSIONS: This study does not support the role of NRG3 in schizophrenia in a Korean population.