ObjectiveTo investigate the current status of depression in women in the second and third trimesters of pregnancy, analyze the factors associated with depressive symptoms during this period,and construct an artificial neural network-based risk prediction model for depressive symptoms, with the aim of providing evidence for early screening, risk identification, and targeted intervention.MethodsA convenience sample of 588 women in the second and third trimesters of pregnancy (28 -36 weeks of pregnancy) who had regular prenatal check-ups in the obstetric outpatient clinics of two tertiary hospitals in Changsha and Hengyang from April 2023 to December 2023 was investigated using the questionnaire for depressive symptoms in the second and third trimesters of pregnancy.Spearman correlation analysis was used to explore the correlation between social support, coping style, vulnerable personality, parenting competence and the Edinburgh postpartum depression score. Binary logistic regression and a multi-layer perceptron artificial neural network were used to construct prediction models for depressive symptoms in the second and third trimesters, respectively.ResultsThe positive rate of depressive symptoms screening in the second and third trimesters was 42.7%.The level of social support, active coping and reactive personality were negatively correlated with depressive symptoms in the second and third trimesters of pregnancy (P< 0.05), while negative coping and vulnerable personality were positively correlated with depressive symptoms in the second and third trimesters of pregnancy (P< 0.05). The prediction accuracy of the logistic regression model for predicting depressive symptoms in the second and third trimesters of pregnancy was 79.6%, and the prediction accuracy of the risk prediction model for depressive symptoms in the second and third trimesters of pregnancy based on artificial neural networks was 86.9%.ConclusionThe positive rate of depression symptom screening is higher in women during the second and third trimesters. Nursing measures can be formulated based on the associated factors in this model to support early identification and targeted intervention for depressive symptoms in the second and third trimesters of pregnancy. The prediction model for depression symptoms in the second and third trimester constructed in this study has good predictive performance and can provide a basis for clinical nurses to identify women at high risk of depressive symptoms.