A predictive model for postpartum depression: ensemble learning strategies in machine learning

Background: Postpartum physical and emotional health remains unaddressed by researchers, clinicians and women themselves. An important factor that affects postpartum recovery is the method of delivery. The objectives of this study were to estimate the incidence of postpartum physical health problems and depression in the first three months after delivery and to identify their relationship with mode of delivery along with other factors associated with the incidence of postpartum depression (PPD).Methods: In this prospective cohort study, women were recruited from three public hospitals within 24-48 hours after delivery. Edinburgh postpartum depression scale (EPDS) and structured validated questionnaires were used. Follow-ups were conducted at 1-week and 3-months after delivery.Results: The most prevalent 3-month physical health problems were insomnia (75%), fatigue (70%), and back pain (65%). There was a high incidence of nipple pain (73%) and vaginal itch (25%) in the spontaneous vaginal delivery (SVD) group during the immediate (1-week) postpartum period, whereas wound pain washigher in the cesarean section (CS) group during the immediate (94%) and 3-month follow-up (55%) postpartum period. The incidence of 3-month PPD (EPDS>=12) was 28% in SVD group and 24% in CS group. There was no effect of the delivery mode on EPDS scores (p=0.59). Wanted pregnancy (OR=0.507), higher number of living children (OR=0.631), gravidity (OR=1.36), and income (OR= 0.536) were protective factors for immediate PPD.Conclusions: Postpartum physical health problems and depression are common after delivery. Health care providers should be aware of these to perform early screening and interventions to alleviate them.

Postpartum depression is a depression syndrome that occurs in mothers after childbirth and can be prevented and cured. According to Fazraningtyas, in South Kalimantan, to be precise in the city of Banjarmasin, namely Ulin General Hospital of Banjarmasin and Dr. H.M. Ansari Saleh General Hospital of Banjarmasin showed 56.8% mild postpartum depression, 26.1% moderate postpartum depression, 17.0% severe postpartum depression. The impact of mothers experiencing postpartum depression is that mothers have difficulty interacting and can endanger their children. Postpartum depression is caused by several factors, the factors that contributed are complications in pregnancy. This study used a literature review approach. The articles obtained from Google Scholar, Biomed Central, and Pubmed. The criterias applied be restricted. As many as 10 journals are found. Based on the previous study, complications in pregnancy are the cause of postpartum depression. Complications that are often experienced by pregnant women are maternal anemia that can appear during the pregnancy process until the birth process occurs and 30-70% of pregnant women with maternal anemia have a risk of postpartum depression. Second, gestational diabetes is one of the complications of pregnancy that occurs in women who are pregnant. Pregnant women can increase hormones including the progesterone hormone, human placental lactogen estrogen, and cortisol. The last, a history of depression is a cause of postpartum depression because if pregnant women have a history of depression before pregnancy, they will have a higher risk of experiencing postpartum depression.

With the emergence of the massive educational data, education data mining techniques have extensively drawn considerable interest from scholars to explore the relationship between students’ achievements and other factors. In this study, the data set about the students’ achievements of Portuguese in two secondary education schools in Portugal is selected for education data mining, which involves the personal information, social and school related factors. To analyze the relationship between the students' achievements and other factors, this study proposed an ensemble model based on weighted voting for predicting the students’ achievements of Portuguese in the final period. First, the raw data is preprocessed using some basic methods, including dummy coding, correlation analysis, standardization, and normalization. Second, the isolation forest algorithm-based outlier adaption is applied to deal with the data set to enhance the robustness of the ensemble model. Finally, two base classifiers, i.e. gradient boosting decision tree and extreme gradient boosting, are integrated to form the ensemble model. The experiments are presented for verifying the superiority of the proposed model by comparing with five base classifiers, including gradient boosting decision tree, adaptive boosting, extreme gradient boosting, random forest, and decision tree. The experimental results demonstrate that the ensemble model performs better than other base classifiers in classification, and prove the validity of the outlier adaption based on isolation forest algorithm.

While past research has linked cannabis use in pregnancy with a history of depression, sparse literature exists on cannabis use during pregnancy and postpartum depression (PPD). In this study, we aimed to better understand the association between PPD and cannabis use during pregnancy in those with and without a history of depression. This was a retrospective cohort study of patients who received prenatal care at a single institution between January 2017 and December 2019. Patient demographics, obstetric history, depression history, substance use history, and Edinburgh Postnatal Depression Scale (EPDS) scores were extracted from patients' medical records. Modified Poisson Regression with robust standard errors was used to estimate the relative risk (RR) of screening positive for PPD, adjusting for age at delivery, race/ethnicity, insurance type, marital status, and smoking history. Among the 799 subjects meeting inclusion criteria, 15.9% used cannabis during pregnancy. There was an increased risk of screening positive for PPD among prenatal cannabis users compared to non-users (aRR = 1.60, 95% CI: (1.05, 2.45)). Among individuals with a history of depression, the adjusted relative risk of screening positive for symptoms of PPD at the postpartum visit was 1.62 times greater in cannabis users compared to non-users (95% CI: (1.02, 2.58)). Prenatal cannabis use is associated with screening positive for PPD, particularly in those individuals with a history of depression. These results should discourage women with depression from self-medicating with cannabis in pregnancy and provide additional support to the existing recommendations to abstain from prenatal cannabis use.

Simultaneous and non-destructive prediction of multiple internal quality characteristics in mandarin citrus with near-infrared spectroscopy and ensemble learning strategy

Postpartum depression (PPD) is a devastating mental illness associated with adverse health outcomes for mother, child, and family. Higher PPD prevalence in First American women suggests a racial/ethnic disparity, yet little is known about how PPD is experienced from their perspective. To 1) describe First American women's PPD experiences and the meanings they ascribe to those experiences and 2) describe the cultural knowledge, influences, and practices during the perinatal period. This phenomenological study used a community-based participatory research approach. Criterion and snowball sampling captured First American women who had PPD now or in the past (N = 8). Interviews used a semistructured guide and thematic analysis followed. Mean age was 30.25 years. Most women were multigravidas (n = 7) and rated themselves as "very" (n = 4) or "mostly" (n = 3) Native American. Women were mostly of low socioeconomic status and had a history of depression (n = 7) and/or a history of prenatal depression (n = 6). Themes: 1) stressors that contributed to PPD; 2) how PPD made me feel; 3) what made my PPD better; 4) heritage-centered practices; 5) support through PPD; 6) how I felt after PPD; and 7) am I a good mother? This study provides a better understanding of some First American women's PPD experiences that facilitates judgment of the importance of PPD within a cultural context. Clinicians need to create culturally appropriate responses to First American women's PPD needs.

Whether the mode of birth delivery correlates with the incidence of postpartum depression (PPD) is still under debate. This study seeks to clarify such a correlation and to further investigate if there are any differences in the likelihood of PPD medical care use among women with and without a history of depression. A logistic regression at two assessment points (6-month and 12-month postpartum) on the National Health Insurance Database in Taiwan is performed. In total, 32,729 women were included who gave first birth from 2007 to 2011 via cesarean section (CS), elective CS, and vaginal delivery, of whom 3580 (10.9%) were diagnosed with a history of depression. Findings show that CS was associated with a higher likelihood of PPD doctor visits regardless of whether the women have a history of depression or not, but elective CS tended to have different impacts for these two groups of women. Mentally healthy women who experienced elective CS had 1.36- and 1.64-times higher risk of PDD medical care use than those who delivered vaginally, whereas previous depressive women undertaking an elective cesarean birth had no significant difference observed in incidences. A notably higher risk of elective CS delivery versus vaginal delivery for mentally healthy women suggests that elective CS is not clinically appropriate, yet it might be an alternative to vaginal delivery with careful counseling for pregnant women who experience a history of depression.

Although the goal of rectal cancer treatment is to restore gastrointestinal continuity, some patients with rectal cancer develop a permanent stoma (PS) after sphincter-saving operations. Although many studies have identified the risk factors and causes of PS, few have precisely predicted the probability of PS formation before surgery. To validate whether an artificial intelligence model can accurately predict PS formation in patients with rectal cancer after sphincter-saving operations. Patients with rectal cancer who underwent a sphincter-saving operation at Taipei Medical University Hospital between January 1, 2012, and December 31, 2021, were retrospectively included in this study. A machine learning technique was used to predict whether a PS would form after a sphincter-saving operation. We included 19 routinely available preoperative variables in the artificial intelligence analysis. To evaluate the efficiency of the model, 6 performance metrics were utilized: accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiving operating characteristic curve. In our classification pipeline, the data were randomly divided into a training set (80% of the data) and a validation set (20% of the data). The artificial intelligence models were trained using the training dataset, and their performance was evaluated using the validation dataset. Synthetic minority oversampling was used to solve the data imbalance. A total of 428 patients were included, and the PS rate was 13.6% (58/428) in the training set. The logistic regression (LR), Gaussian Naïve Bayes (GNB), Extreme Gradient Boosting (XGB), Gradient Boosting (GB), random forest, decision tree and light gradient boosting machine (LightGBM) algorithms were employed. The accuracies of the logistic regression (LR), Gaussian Naïve Bayes (GNB), Extreme Gradient Boosting (XGB), Gradient Boosting (GB), random forest (RF), decision tree (DT) and light gradient boosting machine (LightGBM) models were 70%, 76%, 89%, 93%, 95%, 79% and 93%, respectively. The area under the receiving operating characteristic curve values were 0.79 for the LR model, 0.84 for the GNB, 0.95 for the XGB, 0.95 for the GB, 0.99 for the RF model, 0.79 for the DT model and 0.98 for the LightGBM model. The key predictors that were identified were the distance of the lesion from the anal verge, clinical N stage, age, sex, American Society of Anesthesiologists score, and preoperative albumin and carcinoembryonic antigen levels. Integration of artificial intelligence with available preoperative data can potentially predict stoma outcomes after sphincter-saving operations. Our model exhibited excellent predictive ability and can improve the process of obtaining informed consent.

Association between family members and risk of postpartum depression in Japan: Does “who they live with” matter? -The Japan environment and Children's study

This study conducts an extensive comparative analysis of computational intelligence approaches aimed at predicting the compressive strength (CS) of concrete, utilizing two non-destructive testing (NDT) methods: the rebound hammer (RH) and the ultrasonic pulse velocity (UPV) test. In the ensemble learning approach, the six most popular algorithms (Adaboost, CatBoost, gradient boosting tree (GBT), random forest (RF), stacking, and extreme gradient boosting (XGB)) have been used to develop the prediction models of CS of concrete based on NDT. The ML models have been developed using a total of 721 samples, of which 111 were cast in the laboratory, 134 were obtained from in-situ testing, and the other samples were gathered from the literature. Among the three categories of analytical models—RH models, UPV models, and combined RH and UPV models; seven, ten, and thirteen models have been used respectively. AdaBoost, CatBoost, GBT, RF, Stacking, and XGB models have been used to improve the accuracy and dependability of the analytical models. The RH-M5, UPV-M6, and C-M6 (combined UPV and RH model) models were found with highest performance level amongst all the analytical models. The MAPE value of XGB was observed to be 84.37%, 83.24%, 77.33%, 59.46%, and 81.08% lower than AdaBoost, CatBoost, GBT, RF, and stacking, respectively. The performance of XGB model has been found best than other soft computing techniques and existing traditional predictive models.

Background: Postpartum depression is one of the emotional disturbances that results from failure to postpartum psychological adaptation process. The global prevalence of postpartum depression is 10-15%. This study aimed to analyze the risk factors of pregnancy on the incidence of postpartum depression.
\n
\nSubjects and Method: This was an analytic and observational study with a cross-sectional design. The study was conducted at Dr. Moewardi hospital, from December 2017 to January 2018. A total sample of 150 postpartum mothers was selected for this study by exhaustive sampling. The dependent variable was postpartum depression (PPD). The independent variables were maternal age, stress, parity, unwanted pregnancy, type of labor, family income, family support, and domestic violence. The data were collected by questionnaire and analyzed by path analysis model.
\n
\nResults: Postpartum depression was directly and positively affected by psychological stress (b= 2.15; 95% CI= 1.17 to 3.13; p<0.001), delivery type (b= 1.27; 95% CI= 0.32 to 2.21; p= 0.008), and unwanted pregnancy (b= 1.57; 95% CI= 0.57 to 2.58; p= 0.002). Postpartum depression was directly but negatively affected by family income (b= -1.52; 95% CI= -2.51 to -0.54; p= 0.002), parity (b= -1.24; 95% CI= -2.21 to -0.28; p= 0.011), and family support (b= -1.31; 95% CI= -3.28 to -0.24; p= 0.016). Psychological stress increased with domestic violence (b= 2.68; 95% CI= 0.64 to 4.73; p= 0.010) and decreased with maternal age (b= -0.91; 95% CI= -1.68 to -0.13; p= 0.022). Parity increased with maternal age (b= 1.66; 95% CI= 0.79 to 2.53; p<0.001). Domestic violence decreased with maternal age (b= -1.34; 95% CI= -2.85 to 0.16; p= 0.081).
\n
\nConslusion: Psychological stress, delivery type, unwanted pregnancy, family income, parity, and family support, are direct risk factors for postpartum depression.

Hormonal sensitivity may contribute to the risk of depression in some women, as observed during the premenstrual, postpartum, and perimenopausal phases, and when initiating hormonal contraception (HC). However, little evidence exists to support that such depressive episodes are linked across the reproductive life span. To determine whether prior depression associated with HC initiation is coupled with a higher risk of postpartum depression (PPD) than prior depression not associated with HC initiation. This cohort study used Danish health registry data collected from January 1, 1995, through December 31, 2017, and analyzed from March 1, 2021, through January 1, 2023. All women living in Denmark born after 1978 with their first delivery between January 1, 1996, and June 30, 2017, were eligible for inclusion; 269 354 met these criteria. Women were then excluded if they had never used HC or if they had a depressive episode before 1996 or within 12 months prior to delivery. Prior depression associated with vs not associated with HC initiation, ie, if developed within 6 months after start of an HC exposure or not. Depression was defined as a hospital diagnosis of depression or filling a prescription for antidepressant medication. Crude and adjusted odds ratios (ORs) were calculated for the incidence of PPD defined as the development of depression within 6 months after first delivery. Of 188 648 first-time mothers, 5722 (3.0%) (mean [SD] age, 26.7 [3.9] years) had a history of depression associated with initiation of HC use, and 18 431 (9.8%) (mean [SD] age, 27.1 [3.8] years) had a history of depression not associated with the initiation of HC. Women with HC-associated depression had a higher risk of PPD than women with prior non-HC-associated depression (crude OR, 1.42 [95% CI, 1.24-1.64]; adjusted OR, 1.35 [95% CI, 1.17-1.56]). These findings suggest that a history of HC-associated depression may be associated with a higher risk of PPD, supporting that HC-associated depression may indicate PPD susceptibility. This finding offers a novel strategy in clinical PPD risk stratification and points to the existence of a hormone-sensitive subgroup of women.

Modeling of necking area reduction of carbon steel in hydrogen environment using machine learning approach

Our objectives were to examine the prevalence and incidence of postpartum depressive, anxiety, and comorbid symptoms over the first postpartum year; the persistence of these symptoms; and the prevalence stratified by immigration status. We conducted a longitudinal cohort study in Ontario, Canada. Participants were 571 Chinese recent immigrant, nonrecent immigrant, and Canadian-born women with live births in 2011 to 2014. Participants were assessed at 4, 12, and 52 weeks postpartum for the presence of possible and high depressive symptomatology (Edinburgh Postnatal Depression Scale [EPDS] >9 and >12, respectively), anxiety symptomatology (State-Trait Anxiety Inventory [STAI] >40), and comorbid symptomatology (EPDS >9 and STAI >40). Prevalence and incidence with 95% confidence intervals were calculated. Prevalence rates were highest at 4 weeks and decreased across time, with possible depressive symptomatology most prevalent at most time points. Incidence rates at 12 and 52 weeks were generally similar. Of those with possible symptomatology at 4 weeks, 42.0% or less continued to have symptomatology at 12 weeks and 17.4% or less at 52 weeks. There were no differences in prevalence of any type of symptomatology between immigrant and Canadian-born Chinese women at 4 weeks, but at 12 and 52 weeks, most types of symptomatology were more common among recent and nonrecent immigrants. Our findings suggest that Chinese immigrant women are a high-risk group for postpartum depressive and anxiety symptomatology. Future research should identify cultural and psychosocial factors associated with immigration that could be addressed in the system of care for postpartum immigrant women.

ABSTRACTIn this descriptive study, our aim was to assess the effects of traditional practices on postpartum depression. The participants were 461 Muslim mothers with infants aged between 2 weeks and 12 months, without a history of postpartum depression or other psychiatric problems and capable of communicating fluently. The participants were volunteers and were recruited from family health centers in urban Turkey between January 15 and April 15, 2015. Data were collected in a single face-to-face interview with each participant. Mothers in the research locality frequently have recourse to traditional practices in the postpartum period and the incidence of postpartum depression appears to be lower amongst those who use such practices.