Explore the vast range of titles available.

Postpartum depression (PPD) affects up to 20% of mothers globally. Early detection is vital for better outcomes, yet screening lacks scalability and predictive power. Artificial intelligence (AI)-through machine learning, deep learning, and natural language processing-enhances the early identification of mothers at risk with greater accuracy. This study aims to systematically map the existing literature on AI-based methods for detecting and predicting PPD. This scoping review was conducted in accordance with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines. We included empirical studies that applied AI techniques to detect or predict PPD and were published in peer-reviewed journals, conference proceedings, or dissertations. Studies were excluded if they were nonempirical (eg, reviews, editorials, and abstracts), not published in English, focused on general perinatal mental health without a specific emphasis on PPD, or used AI solely for monitoring or treatment rather than prediction or detection. We systematically searched 8 databases-MEDLINE, Embase, PsycINFO, CINAHL, Scopus, IEEE Xplore, ACM Digital Library, and Google Scholar-from inception through February 28, 2025. The search strategy was supplemented by backward and forward reference screening and biweekly alerts to capture newly published studies. Two independent (M [Alkhateeb] and A [Nayeem])reviewers (M [Alkhateeb] and A [Nayeem]) screened the retrieved studies, with disagreements resolved by a third reviewer (AA [Alrazaq]). Data were extracted by 2 independent reviewers using a standardized extraction form capturing study characteristics, AI model types, data sources, features, preprocessing, validation strategies, and performance metrics. A formal risk-of-bias assessment was not performed due to the scoping nature of the review. All extracted data were synthesized narratively. Out of 503 retrieved studies, 65 met the inclusion criteria. The United States contributed the largest proportion of studies (18/65, 27.7%). The highest number of publications occurred in 2024 (17/65, 26%). Most included studies were journal articles (46/65, 71%). Short-term postpartum outcomes (≤12 weeks) were most frequently assessed (20/65, 30.8%). Most included studies (52/65, 80%) applied AI models for predicting PPD, while 14 of 65 (22%) studies used them for detection. Sociodemographic data were most frequently used (49/65, 75.4%), followed by psychological data (44/65, 68%) and obstetric data (35/65, 55%). Data preprocessing mostly relied on basic scaling (51/65, 79%) and some missing data imputation (29/65, 44.6%). Machine learning dominated (57/65, 87.7%), especially random forest, support vector machines, and logistic regression. Internal validation (k-fold, hold-out) was standard, while external validation was scarce. Ensemble-based boosting models consistently demonstrated superior performance across key metrics, highlighting their potential for accurate and scalable PPD prediction. Current studies suffer from limited sample sizes, geographic bias, lack of standardized feature sets, minimal external validation, and inconsistent reporting of comprehensive model metrics. This scoping review analyzes 65 studies on AI in PPD, highlighting dominant use of classical machine learning, limited deep learning adoption, underuse of advanced preprocessing, inconsistent validation, and reliance on structured, unimodal data-mainly sociodemographic, clinical, and obstetric features.

Postpartum depression (PPD), now referred to as perinatal depression, is a prevalent and debilitating mood disorder that reduces health-related quality of life (HRQoL) and psychosocial functioning. Esketamine, which is efficacious in adults with treatment-resistant depression and individuals with depression and suicidality, is also analgesic in pain management during childbirth labour. Herein, we investigate the efficacy of prophylactic esketamine in reducing the incidence of PPD. We performed a systematic review (i.e., PubMed, Scopus, and Ovid databases; inception to January 22, 2024) of randomized controlled trials that investigated the use of esketamine for PPD. We delimited our search to studies that prespecified the prevention of PPD with esketamine as the primary outcome. A meta-analysis was performed on PPD incidence rates using a random effects model. Our analysis consisted of seven studies that met our eligibility criteria. We found that esketamine was significantly associated with a decreased incidence of PPD diagnosis within one week of childbirth (OR = 0.30, 95% CI = [0.15, 0.60], p = 0.0047). We also observed that esketamine was significantly associated with a decreased incidence of PPD diagnosis between 4 to 6 weeks post-delivery (OR = 0.33, 95% CI = [0.18, 0.59], p = 0.0034). Our results indicate that esketamine may have preventive antidepressant effects during the postpartum period. The aforementioned points have both mechanistic and clinically meaningful implications for the treatment of PPD.

The perinatal period is a vulnerable time when women are at increased risk of depression. \"Mamma Mia\" is a universal preventive internet-delivered intervention offered to pregnant women, with the primary goals of preventing the onset or worsening of depression and enhancing subjective well-being during the perinatal period. However, treatment dropout from internet-delivered interventions is often reported. The study aim was to acquire an understanding of the different experiences among participants who dropped out of the Mamma Mia intervention during pregnancy, compared to participants who dropped out during the postpartum follow-up phase. A total of 16 women from a larger randomized controlled trial (Mamma Mia) participated in individual semistructured interviews following a strengths, weaknesses, opportunities, and threats format. Of the 16 participants included, 8 (50%) women dropped out early from the intervention during pregnancy (pregnancy group), whereas 8 (50%) women dropped out later, after giving birth (postpartum follow-up group). Data were analyzed using the framework approach. The results showed that there were differences between the groups. In general, more participants in the postpartum follow-up group reported that the program was user-friendly. They became more aware of their own thoughts and feelings and perceived that the program had provided them with more new knowledge and practical information than participants in the pregnancy group. Participants in both groups suggested several opportunities for improving the program. There were differences between women who dropped out of the intervention during pregnancy and the postpartum follow-up phase. The reported differences between groups should be further examined.

There is increasing interest in the development of scalable digital mental health interventions for perinatal populations to increase accessibility. Mobile behavioral activation (BA) is efficacious for the treatment of perinatal depression; however, the effect of comorbid anxiety and depression (CAD) on symptom trajectories remains underexplored. This is important given that at least 10% of women in the perinatal period experience CAD. We assessed whether there were differences in symptom trajectories in pregnant participants with CAD as compared to those with depression only (ie, major depressive disorder [MDD]) during intervention with a BA mobile gaming app. Pregnant adults with either CAD (n=10) or MDD (n=7) used a BA app for 10 weeks and completed biweekly symptom severity questionnaires for depression and anxiety. We assessed whether baseline diagnoses were associated with differential symptom trajectories across the study with mixed effects longitudinal models. When controlling for baseline symptoms, results revealed a significant interaction between baseline diagnosis and the quadratic component of study week on anxiety (β=.18, SE 0.07; t62=2.61; P=.01), revealing a tendency for anxiety in the CAD group to increase initially and then decrease at an accelerated rate, whereas MDD symptoms were relatively stable across time. There was a significant effect of linear time on depression (β=-.39, SE 0.11; t68=-3.51; P=.001), showing that depression declined steadily across time for both groups. There was a significant effect of baseline diagnosis on depression (β=-8.53, SE 3.93; t13=-2.17; P=.05), suggesting that those with MDD had higher follow-up depression compared to those with CAD when holding other predictors constant. The app was beneficial in reducing depression symptoms in perinatal individuals with different comorbidity profiles. With respect to anxiety symptom trajectories, however, there was more variability. The app may be especially effective for the treatment of anxiety symptoms among individuals with CAD, as it encourages in-the-moment ecologically relevant exposure to anxiety-provoking stimuli. Despite no significant group difference in baseline anxiety symptoms, the MDD group did not have a significant reduction in their anxiety symptoms across the study period, and some individuals had an increase in anxiety. Findings may point to opportunities for the augmentation of BA gaming apps for those with MDD to more effectively target anxiety symptoms. Overall, findings suggest there may be value in considering comorbidities and individual variations in participants when developing scalable mobile interventions for perinatal populations.

The Edinburgh Postnatal Depression Scale (EPDS) is a widely used screening tool for perinatal depression. Its factor structure is still a debatable topic. Our study aimed to examine the factor structure and measurement invariances of the Japanese version of the EPDS from late pregnancy to early postpartum. A total of 633 women were followed with the EPDS at three times over the perinatal period: late pregnancy (n = 633), 5 days after childbirth (n = 445), and 1 month after childbirth (n = 392). We randomly divided the participants into two groups: one for exploratory factor analyses (EFAs) and another for confirmatory factor analyses (CFAs). The result of the EFAs indicated different factor models at each time point. Hence, CFAs were performed using the second sample set to compare different models including the ones previously reported. A 3-factor model consisting of depression (items 7, 9), anxiety (items 4, 5), and anhedonia (items 1, 2) (Kubota et al., 2018) was consistently stable during the whole perinatal period. Kubota’s 3-factor model showed invariance across the perinatal period.

Background Pre- and post-partum depression is a common mood disorder with detrimental effects on both mother and child. The aim of the proposed review is to summarize evidence related to the effects of both pre- and post-partum depression on child behavior and development from birth to preschool age. In particular, our review will address mutual relations between pre- and post-partum depression in order to determine whether pre- and post-partum depression predict child psychological outcomes independently, whether there is an effect of timing of depression on child outcomes, whether pre- and post-partum depression interact to affect child outcomes, and whether the effect of pre-partum depression is mediated by depression after child’s birth. Methods We will include prospective longitudinal studies that report data about the effects of both pre- and post-partum depression on child psychological outcomes as published in peer-reviewed academic journals since January 1998. We will search EMBASE, MEDLINE, PsycARTICLES, PsycINFO, ISI Web of Science, Scopus, and Wiley Online databases to identify original research articles written in English. Two independent reviewers will screen search results in two stages: (i) titles and abstracts and (ii) full text. The first one will extract data into tables, while the latter will verify whether the data extracted are correct. We will assess the risk of bias in the selected studies using the Critical Appraisal Skills Programme (CASP), Cohort Study Checklist. The results of the review will be reported in a narrative form. If there are sufficient data available, a meta-analysis will be conducted using metaSEM package in R. Discussion The proposed review will be the first systematic review summarizing the effects of both pre- and post-partum depression on child psychological development and behavior from birth to preschool age. The results of such a review may contribute to a better understanding of mutual relations between pre- and post-partum depression in their effects on child outcomes. They may also shed light on what periods in early human development are most vulnerable to the effects of maternal depression. Trial registration PROSPERO CRD42018106269

Despite high rates of perinatal depression and anxiety, little is known about how Aboriginal women in Australia experience these disorders and the acceptability of current clinical screening tools. In a 2014 study, the Kimberley Mum’s Mood Scale (KMMS) was validated as an acceptable perinatal depression and anxiety screening tool for Aboriginal women in the Kimberley region of Western Australia. In the current study, we explored if it was appropriate to trial and validate the KMMS with Aboriginal women in the Pilbara. Yarning as a methodology was used to guide interviews with 15 Aboriginal women in the Pilbara who had received maternal and child health care within the last three years. Data were analysed thematically, the results revealing that this cohort of participants shared similar experiences of stress and hardship during the perinatal period. Participants valued the KMMS for its narrative-based approach to screening that explored the individual’s risk and protective factors. While support for the KMMS was apparent, particular qualities of the administering health care professional were viewed as critical to the tool being well received and culturally safe. Building on these findings, we will work with our partner health services in the Pilbara to validate the KMMS with Pilbara Aboriginal women.

Purpose This systematic review of systematic reviews aims to provide the first global picture of the prevalence and correlates of perinatal depression, and to explore the commonalities and discrepancies of the literature. Methods Seven databases were searched from inception until April 2022. Full-text screening and data extraction were performed independently by two researchers and the AMSTAR tool was used to assess the methodological quality. Results 128 systematic reviews were included in the analysis. Mean overall prevalence of perinatal depression, antenatal depression and postnatal depression was 26.3%, 28.5% and 27.6%, respectively. Mean prevalence was significantly higher (27.4%; SD = 12.6) in studies using self-reported measures compared with structured interviews (17.0%, SD = 4.5; d  = 1.0) and among potentially vulnerable populations (32.5%; SD = 16.7, e.g. HIV-infected African women) compared to the general population (24.5%; SD = 8.1; d  = 0.6). Personal history of mental illness, experiencing stressful life events, lack of social support, lifetime history of abuse, marital conflicts, maternity blues, child care stress, chronic physical health conditions, preeclampsia, gestational diabetes mellitus, being exposed to second-hand smoke and sleep disturbance were among the major correlates of perinatal depression. Conclusion Although the included systematic reviews were all of medium–high quality, improvements in the quality of primary research in this area should be encouraged. The standardisation of perinatal depression assessment, diagnosis and measurement, the implementation of longitudinal designs in studies, inclusions of samples that better represent the population and better control of potentially confounding variables are encouraged.

Background Perinatal depression in women is associated with high morbidity and mortality, and has attracted increasing attention. The investigation of risk factors of perinatal depression in women may contribute to the early identification of depressed or depression-prone women in clinical practice. Material and Methods A computerized systematic literature search was made in Cochrane Library, PubMed, Web of Science, and EMBASE from January 2009 to October 2021. All included articles were published in English, which evaluated factors influencing perinatal depression in women. Based on the recommendations of the Cochrane Collaboration protocols, Review Manager 5.3 was used as a statistical platform. Results Thirty-one studies with an overall sample size of 79,043 women were included in the review. Educational level ( P  = 0.0001, odds ratio [OR]: 1.40, 95% CI: [1.18,1.67]), economic status of families ( P  = 0.0001, OR: 1.69, 95%CI: [1.29,2.22]), history of mental illness ( P  < 0.00001, OR: 0.29, 95% CI: [0.18, 0.47]), domestic violence ( P  < 0.00001, OR: 0.24, 95% CI: [0.17,0.34]), perinatal smoking or drinking ( P  = 0.005, OR: 0.63; 95% CI [0.45, 0.87]; P  = 0.008, OR: 0.43, 95% CI, [0.23 to 0.80]; respectively), and multiparity( P  = 0.0003, OR: 0.74, 95% CI: [0.63, 0.87]) were correlated with perinatal depression in women. The stability of our pooled results was verified by sensitivity analysis and publication bias was not observed based on funnel plot results. Conclusion Lower educational level, poor economic status of families, history of mental illness, domestic violence, perinatal smoking or drinking, and multiparity serve as risk factors of perinatal depression in women.

Depression during pregnancy and after the birth of a child is highly prevalent and an important public health problem. Psychological interventions are the first-line treatment and, although a considerable number of randomized trials have been conducted, no recent comprehensive meta-analysis has evaluated treatment effects. We used an existing database of randomized controlled trials of psychotherapies for adult depression and included studies aimed at perinatal depression. Random effects models were used in all analyses. We examined the effects of the interventions in the short and long term, and also examined secondary outcomes. Forty-three studies with 49 comparisons and 6270 participants between an intervention and control group were included. The overall effect size was = 0.67 [95% confidence interval (CI) 0.45~0.89; numbers needed-to-be-treated = 4.39] with high heterogeneity ( = 80%; 95% CI 75~85). This effect size remained largely unchanged and significant in a series of sensitivity analyses, although some publication bias was found. The effects remained significant at 6-12 months follow-up. Significant effects were also found for social support, anxiety, functional limitations, parental stress and marital stress, although the number of studies for each outcome was low. All results should be considered with caution because of the high levels of heterogeneity in most analyses. Psychological interventions are probably effective in the treatment of perinatal depression, with effects that last at least up to 6-12 months and probably also have effects on social support, anxiety, functional impairment, parental stress, and marital stress.