A Study on Multimodal Data- and Explainable AI-Based Risk Prediction and Intervention for Postpartum Depression in Nursing Care Settings

Yangyang Lin

doi:10.62177/apjcmr.v2i5.1788

Authors

Yangyang Lin Zhejiang Dongfang Polytechnic

DOI:

https://doi.org/10.62177/apjcmr.v2i5.1788

Keywords:

Postpartum Depression, Multimodal Data, Explainable Artificial Intelligence, Nursing Decision Support, Risk Stratification

Abstract

Objective: To evaluate a nursing-centered dynamic prediction and risk-stratified intervention system for postpartum depression (PPD) using multimodal longitudinal data and explainable artificial intelligence (AI). Methods: A quasi-experimental controlled study included 300 postpartum women enrolled within 6 hours after delivery and allocated by admission sequence to AI-assisted stratified nursing (n=150) or routine nursing care (n=150). Physiological, behavioral, psychological, and social-support indicators were collected, with EPDS assessments on postpartum days 7, 14, 28, and 42. A hybrid model combined temporal convolution, long short-term memory, text and voice representations, and multilayer fusion; SHAP and attention mechanisms supported case-level explanations. Outcomes included time to risk identification, EPDS trajectory, remission among high-risk women, nursing workload, and internally validated model performance. Results: Risk was identified earlier in the intervention group than in the control group (1.21 +/- 0.68 vs. 4.73 +/- 1.92 days, P<0.05). Mean EPDS reduction by day 42 was greater (-5.62 +/- 3.14 vs. -2.13 +/- 2.87, P<0.05), and remission was more frequent (78.3% vs. 42.6%; chi-square=36.41, P<0.05). High-risk identification reached 92.0% versus 64.0%. Time per follow-up decreased by 34.7%, and nurses reported fewer women requiring daily follow-up and lower work-related stress. The model achieved an AUC of 0.864, accuracy of 0.82, recall of 0.77, and F1-score of 0.79. Conclusion: A multimodal explainable AI system integrated with risk-stratified nursing care was associated with earlier risk recognition, improved depressive symptom trajectories, and lower nursing workload. External validation and a fully randomized multicenter evaluation are required before broad implementation.

Downloads

Download data is not yet available.

References

Shorey, S., Chee, C. Y. I., Ng, E. D., Chan, Y. H., Tam, W. W. S., & Chong, Y. S. (2018). Prevalence and incidence of postpartum depression among healthy mothers: A systematic review and meta-analysis. Journal of Psychiatric Research, 104, 235–248. https://doi.org/10.1016/j.jpsychires.2018.08.001

Meltzer-Brody, S., Howard, L. M., Bergink, V., et al. (2018). Postpartum psychiatric disorders. Nature Reviews Disease Primers, 4, Article 18022. https://doi.org/10.1038/nrdp.2018.22

World Health Organization. (2022). Guide for integration of perinatal mental health in maternal and child health services. World Health Organization.

Cox, J. L., Holden, J. M., & Sagovsky, R. (1987). Detection of postnatal depression: Development of the 10-item Edinburgh Postnatal Depression Scale. British Journal of Psychiatry, 150, 782–786. https://doi.org/10.1192/bjp.150.6.782

Levis, B., Negeri, Z., Sun, Y., et al. (2020). Accuracy of the Edinburgh Postnatal Depression Scale for screening to detect major depression among pregnant and postpartum women: Systematic review and meta-analysis of individual participant data. BMJ, 371, Article m4022. https://doi.org/10.1136/bmj.m4022

Qi, W., Wang, Y., Li, C., et al. (2023). Predictive models for predicting the risk of maternal postpartum depression: A systematic review and evaluation. Journal of Affective Disorders, 333, 107–120. https://doi.org/10.1016/j.jad.2023.04.026

Shin, D., Lee, K. J., Adeluwa, T., & Hur, J. (2020). Machine learning-based predictive modeling of postpartum depression. Journal of Clinical Medicine, 9(9), Article 2899. https://doi.org/10.3390/jcm9092899

Lin, Y., & Zhou, D. (2025). A method for predicting postpartum depression via an ensemble neural network model. Frontiers in Public Health, 13, Article 1571522. https://doi.org/10.3389/fpubh.2025.1571522

Lilhore, U. K., Dalal, S., Varshney, N., et al. (2024). Prevalence and risk factors analysis of postpartum depression at early stage using a hybrid deep learning model. Scientific Reports, 14, Article 4533. https://doi.org/10.1038/s41598-024-54927-8

Hurwitz, E., Butzin-Dozier, Z., Master, H. T., et al. (2024). Harnessing consumer wearable digital biomarkers for individualized recognition of postpartum depression using the All of Us Research Program data set: Cross-sectional study. JMIR mHealth and uHealth, 12, Article e54622. https://doi.org/10.2196/54622

Baltrusaitis, T., Ahuja, C., & Morency, L. P. (2019). Multimodal machine learning: A survey and taxonomy. IEEE Transactions on Pattern Analysis and Machine Intelligence, 41(2), 423–443. https://doi.org/10.1109/TPAMI.2018.2798607

Tonekaboni, S., Joshi, S., McCradden, M. D., & Goldenberg, A. (2019). What clinicians want: Contextualizing explainable machine learning for clinical end use. Proceedings of Machine Learning Research, 106, 359–380.

Collins, G. S., Moons, K. G. M., Dhiman, P., et al. (2024). TRIPOD+AI statement: Updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ, 385, Article e078378. https://doi.org/10.1136/bmj-2023-078378

Desai, P. M., Harkins, S., Rahman, S., et al. (2024). Visualizing machine learning-based predictions of postpartum depression risk for lay audiences. Journal of the American Medical Informatics Association, 31, 289–297. https://doi.org/10.1093/jamia/ocad198

Dennis, C. L., & Dowswell, T. (2013). Psychosocial and psychological interventions for preventing postpartum depression. Cochrane Database of Systematic Reviews, (2). https://doi.org/10.1002/14651858.CD001134.pub3

Topol, E. J. (2019). High-performance medicine: The convergence of human and artificial intelligence. Nature Medicine, 25, 44–56. https://doi.org/10.1038/s41591-018-0300-7

Wiens, J., Saria, S., Sendak, M., et al. (2019). Do no harm: A roadmap for responsible machine learning for health care. Nature Medicine, 25, 1337–1340. https://doi.org/10.1038/s41591-019-0548-6

Liu, H., Dai, A., Zhou, Z., et al. (2023). An optimization for postpartum depression risk assessment and preventive intervention strategy based on machine learning approaches. Journal of Affective Disorders, 328, 163–174. https://doi.org/10.1016/j.jad.2023.028