Interpretable Machine Learning Model for Predicting Postpartum Depression: Retrospective Study
Abstract BackgroundPostpartum depression (PPD) is a prevalent mental health issue with significant impacts on mothers and families. Exploring reliable predictors is crucial for the early and accurate prediction of PPD, which remains challenging. ObjectiveThis study...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
JMIR Publications
2025-01-01
|
| Series: | JMIR Medical Informatics |
| Online Access: | https://medinform.jmir.org/2025/1/e58649 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585079357964288 |
|---|---|
| author | Ren Zhang Yi Liu Zhiwei Zhang Rui Luo Bin Lv |
| author_facet | Ren Zhang Yi Liu Zhiwei Zhang Rui Luo Bin Lv |
| author_sort | Ren Zhang |
| collection | DOAJ |
| description |
Abstract
BackgroundPostpartum depression (PPD) is a prevalent mental health issue with significant impacts on mothers and families. Exploring reliable predictors is crucial for the early and accurate prediction of PPD, which remains challenging.
ObjectiveThis study aimed to comprehensively collect variables from multiple aspects, develop and validate machine learning models to achieve precise prediction of PPD, and interpret the model to reveal clinical implications.
MethodsThis study recruited pregnant women who delivered at the West China Second University Hospital, Sichuan University. Various variables were collected from electronic medical record data and screened using least absolute shrinkage and selection operator penalty regression. Participants were divided into training (1358/2055, 66.1%) and validation (697/2055, 33.9%) sets by random sampling. Machine learning–based predictive models were developed in the training cohort. Models were validated in the validation cohort with receiver operating curve and decision curve analysis. Multiple model interpretation methods were implemented to explain the optimal model.
ResultsWe recruited 2055 participants in this study. The extreme gradient boosting model was the optimal predictive model with the area under the receiver operating curve of 0.849. Shapley Additive Explanation indicated that the most influential predictors of PPD were antepartum depression, lower fetal weight, elevated thyroid-stimulating hormone, declined thyroid peroxidase antibodies, elevated serum ferritin, and older age.
ConclusionsThis study developed and validated a machine learning–based predictive model for PPD. Several significant risk factors and how they impact the prediction of PPD were revealed. These findings provide new insights into the early screening of individuals with high risk for PPD, emphasizing the need for comprehensive screening approaches that include both physiological and psychological factors. |
| format | Article |
| id | doaj-art-89bca0567e3e48e8aa37cefc6cfa54df |
| institution | Kabale University |
| issn | 2291-9694 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | JMIR Publications |
| record_format | Article |
| series | JMIR Medical Informatics |
| spelling | doaj-art-89bca0567e3e48e8aa37cefc6cfa54df2025-01-27T04:38:47ZengJMIR PublicationsJMIR Medical Informatics2291-96942025-01-0113e58649e5864910.2196/58649Interpretable Machine Learning Model for Predicting Postpartum Depression: Retrospective StudyRen Zhanghttp://orcid.org/0009-0002-9469-384XYi Liuhttp://orcid.org/0000-0003-4785-0021Zhiwei Zhanghttp://orcid.org/0009-0003-2146-3901Rui Luohttp://orcid.org/0009-0008-0048-4531Bin Lvhttp://orcid.org/0000-0001-8301-638X Abstract BackgroundPostpartum depression (PPD) is a prevalent mental health issue with significant impacts on mothers and families. Exploring reliable predictors is crucial for the early and accurate prediction of PPD, which remains challenging. ObjectiveThis study aimed to comprehensively collect variables from multiple aspects, develop and validate machine learning models to achieve precise prediction of PPD, and interpret the model to reveal clinical implications. MethodsThis study recruited pregnant women who delivered at the West China Second University Hospital, Sichuan University. Various variables were collected from electronic medical record data and screened using least absolute shrinkage and selection operator penalty regression. Participants were divided into training (1358/2055, 66.1%) and validation (697/2055, 33.9%) sets by random sampling. Machine learning–based predictive models were developed in the training cohort. Models were validated in the validation cohort with receiver operating curve and decision curve analysis. Multiple model interpretation methods were implemented to explain the optimal model. ResultsWe recruited 2055 participants in this study. The extreme gradient boosting model was the optimal predictive model with the area under the receiver operating curve of 0.849. Shapley Additive Explanation indicated that the most influential predictors of PPD were antepartum depression, lower fetal weight, elevated thyroid-stimulating hormone, declined thyroid peroxidase antibodies, elevated serum ferritin, and older age. ConclusionsThis study developed and validated a machine learning–based predictive model for PPD. Several significant risk factors and how they impact the prediction of PPD were revealed. These findings provide new insights into the early screening of individuals with high risk for PPD, emphasizing the need for comprehensive screening approaches that include both physiological and psychological factors.https://medinform.jmir.org/2025/1/e58649 |
| spellingShingle | Ren Zhang Yi Liu Zhiwei Zhang Rui Luo Bin Lv Interpretable Machine Learning Model for Predicting Postpartum Depression: Retrospective Study JMIR Medical Informatics |
| title | Interpretable Machine Learning Model for Predicting Postpartum Depression: Retrospective Study |
| title_full | Interpretable Machine Learning Model for Predicting Postpartum Depression: Retrospective Study |
| title_fullStr | Interpretable Machine Learning Model for Predicting Postpartum Depression: Retrospective Study |
| title_full_unstemmed | Interpretable Machine Learning Model for Predicting Postpartum Depression: Retrospective Study |
| title_short | Interpretable Machine Learning Model for Predicting Postpartum Depression: Retrospective Study |
| title_sort | interpretable machine learning model for predicting postpartum depression retrospective study |
| url | https://medinform.jmir.org/2025/1/e58649 |
| work_keys_str_mv | AT renzhang interpretablemachinelearningmodelforpredictingpostpartumdepressionretrospectivestudy AT yiliu interpretablemachinelearningmodelforpredictingpostpartumdepressionretrospectivestudy AT zhiweizhang interpretablemachinelearningmodelforpredictingpostpartumdepressionretrospectivestudy AT ruiluo interpretablemachinelearningmodelforpredictingpostpartumdepressionretrospectivestudy AT binlv interpretablemachinelearningmodelforpredictingpostpartumdepressionretrospectivestudy |