Detecting risk of suicide attempts among Chinese medical college students using a machine learning algorithm.

BACKGROUND: Suicide has become one of the most prominent concerns for public health and wellness; however, detecting suicide risk factors among individuals remains a big challenge. The aim of this study was to develop a machine learning algorithm that could effectively and accurately identify the probability of suicide attempts in medical college students. METHODS: A total of 4,882 medical students were enrolled in this cross-sectional study. Self-report data on socio-demographic and clinical characteristics were collected online via website or through the widely used social media app, WeChat. 5-fold cross validation was used to build a random forest model with 37 suicide attempt predictors. Model performance was measured for sensitivity, specificity, area under the curve (AUC), and accuracy. All analyses were conducted in MATLAB. RESULTS: The random forest model achieved good performance [area under the curve (AUC) = 0.9255] in predicting suicide attempts with an accuracy of 90.1% (SD = 0.67%), sensitivity of 73.51% (SD = 2.33%) and specificity of 91.68% (SD = 0.82%). LIMITATION: The participants are primarily females and medical students. CONCLUSIONS: This study demonstrates that the random forest model has the potential to predict suicide attempts among medical college students with high accuracy. Our findings suggest that application of the machine learning model may assist in improving the efficiency of suicide prevention.

TSPO upregulation in bipolar disorder and concomitant downregulation of mitophagic proteins and NLRP3 inflammasome activation.

Bipolar disorder (BD) is a chronic, debilitating illness with a global prevalence of up to 4.8%. The importance of understanding how dysfunctional mitochondria and mitophagy contribute to cell survival and death in BD is becoming increasingly apparent. Therefore, the purpose of this study was to evaluate the mitophagic pathway and NLRP3 inflammasome activation in peripheral blood mononuclear cells (PBMCs) of patients with BD and healthy individuals. Since 18-kDa translocator protein (TSPO) plays an important role in regulating mitochondrial function and since TSPO itself impairs cellular mitophagy, we also investigated the changes in the TSPO-related pathway. Our results showed that patients with BD had lower levels of Parkin, p62/SQSTM1 and LC3A and an upregulation of TSPO pathway proteins (TSPO and VDAC), both in terms of mRNA and protein levels. Additionally, we found a negative correlation between mitophagy-related proteins and TSPO levels, while VDAC correlated negatively with p62/SQSTM1 and LC3 protein levels. Moreover, we found that the gene expression levels of the NLRP3-related proteins NLRP3, ASC, and pro-casp1 were upregulated in BD patients, followed by an increase in caspase-1 activity as well as IL-1ß and IL-18 levels. As expected, there was a strong positive correlation between NLRP3-related inflammasome activation and TSPO-related proteins. The data reported here suggest that TSPO-VDAC complex upregulation in BD patients, the simultaneous downregulation of mitophagic proteins and NLRP3 inflammasome activation could lead to an accumulation of dysfunctional mitochondria, resulting in inflammation and apoptosis. In summary, the findings of this study provide novel evidence that mitochondrial dysfunction measured in peripheral blood is associated with BD.