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.

Changes of new coagulation markers in healthy pregnant women and establishment of reference intervals in Changsha / 中南大学学报(医学版)

OBJECTIVES@#There is a high coagulation state in pregnant women, which is prone to coagulation and fibrinolysis system dysfunction. This study aims to explore the latest coagulation markers-thrombomodulin (TM), thrombin-antithrombin complex (TAT), plasmin-α2 plasmin inhibitor complex (PIC), and tissue plasminogen activator/plasminogen activator inhibitor compound (tPAI-C) in different stages of pregnancy, establish reference intervals (RIs) for healthy pregnant women of Chinese population, and to provide an effective and reliable reference for clinicians.@*METHODS@#A total of 492 healthy pregnant women, who underwent pregnancy examination and delivery in the Department of Obstetrics, Second Xiangya Hospital of Central South University from October 2019 to October 2020, were enrolled for this study. They were assigned into the first trimester group, the second trimester group, the third trimester group, and the puerperium group according to the pregnancy period, and 123 healthy non-pregnant women were selected as the controls. Plasma levels of TM, TAT, PIC and tPAI-C were analyzed by automatic chemiluminescence immunoassay analyzer. The RIs for TM, TAT, PIC, and tPAI-C were defined using non-parametric 95% intervals, determined following Clinical and Laboratory Standards Institute Document C28-A3c (CLSI C28-A3c), and Formulation of Reference Intervals for the Clinical Laboratory Test Items (WS/T402-2012).@*RESULTS@#TM and TAT levels increased gradually in the first, second, and third trimester women and decreased in the puerperium women (P<0.05 or P<0.01). PIC level of healthy non-pregnant women was lower than that of pregnant women (P<0.05 or P<0.01), but PIC level of pregnant and puerperium women did not differ significantly (P>0.05). tPAI-C level in healthy non-pregnant women was lower than that of pregnant women (P<0.05 or P<0.01), and tPAI-C level was significantly decreases in the puerperium women (P<0.01). The RIs for TM were as follows: Healthy non-pregnant women at 3.20-4.60 TU/mL, the first and second trimester at 3.12-7.90 TU/mL, the third trimester at 3.42-8.29 TU/mL, puerperium at 2.70-6.40 TU/mL. The RIs for TAT were as follows: Healthy non-pregnant women at 0.50-1.64 ng/mL, the first and second trimester at 0.52-6.91 ng/mL, the third trimester at 0.96-12.92 ng/mL, puerperium at 0.82-3.75 ng/mL. The RIs for PIC were as follows: Healthy non-pregnant women at 0.160-0.519 ng/mL, pregnant women at 0.162-0.770 μg/mL. The RIs for tPAI-C were as follows: Healthy non-pregnant women at 1.90-4.80 ng/mL, the first and second trimester at 2.03-9.33 ng/mL, the third trimester at 2.80-14.20 ng/mL, puerperium at 1.10-8.40 ng/mL.@*CONCLUSIONS@#The levels of 4 new coagulation markers TM, TAT, PIC, and tPAI-C in pregnant women are increased significantly during pregnancy and gradually return to normal after delivery. The RIs for TM, TAT, PIC, and tPAI-C in pregnant women by trimester are established according to CLSI C28-A3c, thus providing a clinical reference for clinician in judgement of thrombotic risk.