Use of artificial intelligence to improve perinatal mental health: a systematic review
Bibliographic record
Abstract
Abstract Introduction Artificial intelligence (AI) technologies have successfully been used in healthcare. Perinatal mental health disorders (PMHD) are detrimental conditions that may affect 10-20% of new mothers. Preventive interventions amongst high-risk women can improve the outcome for mothers and babies. However, our ability to identify these is poor. This systematic review aims to investigate whether AI techniques can improve the early identification of women at high risk of PMHD. Methods This review was conducted applying the PRISMA guidelines. We searched: PubMed, Scopus, Web of Science and Cochrane databases, selecting articles published between 2015 and March 2025. We focused on primary studies, English language. Articles had to report AI technologies applied to improve the early identification of women at high risk of PMHD. Quality of the included studies was assessed using the Newcastle-Ottawa Scale, based on retrospective cohort studies. Results Out of the initial 510 articles identified, 16 were included in the final review. The preliminary results from these studies mostly showed homogeneous findings: AI, particularly machine learning models (eg Random Forest, XGBoost, SVM,LASSO) demonstrated an overall effectiveness in the early detection of PMHD, especially postpartum depression, and improvement in the prediction accuracy compared to conventional regression models, with area under the receiver operator characteristic curve (AUC) within a range of 0.71-0.84, sensitivity of 0,71-0,76 and specificity of 0,75-0,85. These results show that AI models applied to clinical health data enhance prediction accuracy for PMHD at an early stage. Conclusions AI models demonstrated to be promising complementary tools for the early identification of PMHD, improving the performance of standard screening tools. However, variability in study designs, sample populations and algorithms limit the generalizability of the results and highlights the need for further standardized research. Key messages • Consistent evidence shows that AI effectively improves PMHD prediction. • AI can enhance early detection of PMHD, complementing current screening methods and enabling timely intervention.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.011 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".