Using machine learning algorithm to predict the risk of post-traumatic stress disorder among firefighters in Changsha / 中南大学学报(医学版)

OBJECTIVES@#Firefighters are prone to suffer from psychological trauma and post-traumatic stress disorder (PTSD) in the workplace, and have a poor prognosis after PTSD. Reliable models for predicting PTSD allow for effective identification and intervention for patients with early PTSD. By collecting the psychological traits, psychological states and work situations of firefighters, this study aims to develop a machine learning algorithm with the aim of effectively and accurately identifying the onset of PTSD in firefighters, as well as detecting some important predictors of PTSD onset.@*METHODS@#This study conducted a cross-sectional survey through convenient sampling of firefighters from 20 fire brigades in Changsha, which were evenly distributed across 6 districts and Changsha County, with a total of 628 firefighters. We used the synthetic minority oversampling technique (SMOTE) to process data sets and used grid search to finish the parameter tuning. The predictive capability of several commonly used machine learning models was compared by 5-fold cross-validation and using the area under the receiver operating characteristic curve (ROC-AUC), accuracy, precision, recall, and F1 score.@*RESULTS@#The random forest model achieved good performance in predicting PTSD with an average AUC score at 0.790. The mean accuracy of the model was 90.1%, with an F1 score of 0.945. The three most important predictors were perseverance, forced thinking, and reflective deep thinking, with weights of 0.165, 0.158, and 0.152, respectively. The next most important predictors were employment time, psychological power, and optimism.@*CONCLUSIONS@#PTSD onset prediction model for Changsha firefighters constructed by random forest has strong predictive ability, and both psychological characteristics and work situation can be used as predictors of PTSD onset risk for firefighters. In the next step of the study, validation using other large datasets is needed to ensure that the predictive models can be used in clinical setting.

High efficiency of left superior frontal gyrus and the symptom features of major depressive disorder / 中南大学学报(医学版)

OBJECTIVES@#Major depressive disorder (MDD) patients with anhedonia tend to have a poor prognosis. The underlying imaging basis for anhedonia in MDD remains largely unknown. The relationship between nodal properties and anhedonia in MDD patients need to be further investigated. Herein, this study aims to explore differences of cerebral functional node characteristics in MDD patients with severe anhedonia (MDD-SA) and MDD patients with mild anhedonia (MDD-MA) before and after the antidepressant treatment.@*METHODS@#Ninety participants with current MDD were recruited in this study. 24-Item Hamilton Depression Scale (HAMD-24) and Snaith-Hamilton Pleasure Scale (SHAPS) were used to assess the severity of depression and anhedonia at baseline and the end of 6-months treatment. The MDD patients who scored above the 25th percentile on the SHAPS were assigned to an MDD-SA group (n=19), while those who scored below the 25th percentile were assigned to an MDD-MA group (n=18). All patients in the 2 groups received antidepressant treatment. Functional magnetic resonance imaging (fMRI) images of all the patients were collected at baseline and the end of 6-months treatment. Graph theory was applied to analyze the patients' cerebral functional nodal characteristics, which were measured by efficiency (ei) and degree (ki).@*RESULTS@#Repeated measures 2-factor ANCOVA showed significant main effects on group on the ei and ki values of left superior frontal gyrus (LSFG) (P=0.003 and P=0.008, respectively), and on the ei and ki values of left medial orbital-frontal gyrus (LMOFG) (P=0.004 and P=0.008, respectively). Compared with the MDD-MA group, the significantly higher ei and ki values of the LSFG (P=0.015 and P=0.021, respectively), and the significantly higher ei and ki values of the LMOFG (P=0.015 and P=0.037, respectively) were observed in the MDD-SA group at baseline. Meanwhile, higher SHAPS scores could result in higher ei and ki values of LSFG (P=0.019 and P=0.026, respectively), and higher ei value of LMOFG (P=0.040) at baseline; higher SHAPS scores could result in higher ei values of LSFG (P=0.049) at the end of 6-months treatment. The multiple linear regression analysis revealed that sex were negatively correlated with the ei and ki values of LSFG (r= -0.014, P=0.004; r=-1.153, P=0.001, respectively). The onset age of MDD was negatively correlated with the ki value of LSFG (r=-0.420, P=0.034) at the end of 6-months treatment. We also found that SHAPS scores at baseline were positively correlated with the HAMD-24 scores (r=0.387, P=0.022) at the end of 6-months treatment.@*CONCLUSIONS@#There are obvious differences in nodal properties between the MDD-SA and the MDD-MA patients, such as the high ei of LSFG in the MDD-SA patients, which may be associated with the severity of anhedonia. These nodal properties could be potential biomarkers for the prognosis of MDD. The increased ei and ki values in the LSFG of MDD-SA patients may underlie a compensatory mechanism or protective mechanism. The mechanism may be an important component of the pathological mechanism of MDD-SA. The poor prognosis in the MDD-SA patients suggests that anhedonia may predict a worse prognosis in MDD patients. Sex and onset age of MDD may affect the nodal properties of LSFG at baseline and the end of 6-months treatment.

Mental health status of patients with coronavirus disease 2019 in Changsha / 中南大学学报(医学版)

OBJECTIVES@#The epidemic of coronavirus disease 2019 (COVID-19) brought psychological stress to the public, especially to patients. This study aims to investigate the mental health of patients with COVID-19 in Changsha.@*METHODS@#We took cross-section investigation for the mental health of 112 patients with COVID-19 via questionnaires. Mann-Whitney test, Chi-square test, and Fisher's exact test were performed to compare general and clinical data between the slight-ordinary patients and severe patients. Single sample -tests were used to compare the difference between the factor scores of the Symptom Check-List 90 (SCL-90) in COVID-19 patients with the norm of 2015 and factor scores of SCL-90 in patients with the severe acute respiratory syndrome (SARS).@*RESULTS@#The obsessive-compulsive, depression, sleep and eating disorders had the highest frequency among the positive symptoms of SCL-90 in patients with COVID-19 in Changsha. The factor scores of somatization, depression, anxiety, phobia anxiety, sleep and eating disorders in patients with COVID-19 were higher than those of the norm (≤0.001 or 0.05).@*CONCLUSIONS@#The levels of somatization, depression, anxiety, phobia anxiety, sleep and eating disorders in patients with COVID-19 in Changsha are higher than those of the norm. However, the mental health of slight-ordinary patients with COVID-19 is better than that of patients with SARS. It needs to provide targeting psychological interventions depending on the severity of patients.