Integrating SHAP analysis with machine learning to predict postpartum hemorrhage in vaginal births

Abstract Objective This study aimed to develop a machine learning (ML) model integrated with SHapley Additive exPlanations (SHAP) analysis to predict postpartum hemorrhage (PPH) following vaginal deliveries, offering a potential tool for personalized risk assessment and prevention in clinical settin...

Full description

Saved in:
Bibliographic Details
Main Authors: Zixuan Song, Hong Lin, Mengyuan Shao, Xiaoxue Wang, Xueting Chen, Yangzi Zhou, Dandan Zhang
Format: Article
Language:English
Published: BMC 2025-05-01
Series:BMC Pregnancy and Childbirth
Subjects:
Online Access:https://doi.org/10.1186/s12884-025-07633-w
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850206062919024640
author Zixuan Song
Hong Lin
Mengyuan Shao
Xiaoxue Wang
Xueting Chen
Yangzi Zhou
Dandan Zhang
author_facet Zixuan Song
Hong Lin
Mengyuan Shao
Xiaoxue Wang
Xueting Chen
Yangzi Zhou
Dandan Zhang
author_sort Zixuan Song
collection DOAJ
description Abstract Objective This study aimed to develop a machine learning (ML) model integrated with SHapley Additive exPlanations (SHAP) analysis to predict postpartum hemorrhage (PPH) following vaginal deliveries, offering a potential tool for personalized risk assessment and prevention in clinical settings. Methods We conducted a retrospective multicenter cohort study in Northeast China, including women who had vaginal deliveries at three tertiary hospitals from September 2018 to December 2023. Data were extracted from electronic medical records. The dataset was split into a training set (70%) and an internal validation set (30%) to prevent overfitting. External validation was performed on a separate dataset. Several evaluation metrics, including the area under the receiver operating characteristic curve (AUC), were used to compare prediction performance. Features were ranked using SHAP, and the final model was explained. Results The XGBoost model demonstrated superior predictive accuracy for PPH, with an AUC of 0.997 in the training set. SHAP value-based feature selection identified 15 key features contributing to the model’s predictive power. SHAP dependence and summary plots provided intuitive insights into each feature’s contribution, enabling the identification of anomalies. The final model maintained high predictive power, with an AUC of 0.894 in internal validation and 0.880 in external validation. Conclusion This study successfully developed an interpretable ML model that predicts PPH with high accuracy. Future studies with larger and more diverse datasets are necessary to further validate and refine the model, particularly to assess its generalizability across different populations and healthcare settings.
format Article
id doaj-art-e478cd7bbb874c6483d947744a6dd843
institution OA Journals
issn 1471-2393
language English
publishDate 2025-05-01
publisher BMC
record_format Article
series BMC Pregnancy and Childbirth
spelling doaj-art-e478cd7bbb874c6483d947744a6dd8432025-08-20T02:10:56ZengBMCBMC Pregnancy and Childbirth1471-23932025-05-0125111510.1186/s12884-025-07633-wIntegrating SHAP analysis with machine learning to predict postpartum hemorrhage in vaginal birthsZixuan Song0Hong Lin1Mengyuan Shao2Xiaoxue Wang3Xueting Chen4Yangzi Zhou5Dandan Zhang6Department of Obstetrics and Gynecology, Shengjing Hospital of China Medical UniversityDepartment of Obstetrics and Gynecology, Liaoning Maternal and Child Health HospitalDepartment of Obstetrics and Gynecology, Shenyang Women’s and Children’s HospitalDepartment of Health Management, Shengjing Hospital of China Medical UniversityDepartment of Health Management, Shengjing Hospital of China Medical UniversityDepartment of Obstetrics and Gynecology, Shengjing Hospital of China Medical UniversityDepartment of Obstetrics and Gynecology, Shengjing Hospital of China Medical UniversityAbstract Objective This study aimed to develop a machine learning (ML) model integrated with SHapley Additive exPlanations (SHAP) analysis to predict postpartum hemorrhage (PPH) following vaginal deliveries, offering a potential tool for personalized risk assessment and prevention in clinical settings. Methods We conducted a retrospective multicenter cohort study in Northeast China, including women who had vaginal deliveries at three tertiary hospitals from September 2018 to December 2023. Data were extracted from electronic medical records. The dataset was split into a training set (70%) and an internal validation set (30%) to prevent overfitting. External validation was performed on a separate dataset. Several evaluation metrics, including the area under the receiver operating characteristic curve (AUC), were used to compare prediction performance. Features were ranked using SHAP, and the final model was explained. Results The XGBoost model demonstrated superior predictive accuracy for PPH, with an AUC of 0.997 in the training set. SHAP value-based feature selection identified 15 key features contributing to the model’s predictive power. SHAP dependence and summary plots provided intuitive insights into each feature’s contribution, enabling the identification of anomalies. The final model maintained high predictive power, with an AUC of 0.894 in internal validation and 0.880 in external validation. Conclusion This study successfully developed an interpretable ML model that predicts PPH with high accuracy. Future studies with larger and more diverse datasets are necessary to further validate and refine the model, particularly to assess its generalizability across different populations and healthcare settings.https://doi.org/10.1186/s12884-025-07633-wPostpartum hemorrhageVaginal birthsInterpretable modelXGBoostSHAP
spellingShingle Zixuan Song
Hong Lin
Mengyuan Shao
Xiaoxue Wang
Xueting Chen
Yangzi Zhou
Dandan Zhang
Integrating SHAP analysis with machine learning to predict postpartum hemorrhage in vaginal births
BMC Pregnancy and Childbirth
Postpartum hemorrhage
Vaginal births
Interpretable model
XGBoost
SHAP
title Integrating SHAP analysis with machine learning to predict postpartum hemorrhage in vaginal births
title_full Integrating SHAP analysis with machine learning to predict postpartum hemorrhage in vaginal births
title_fullStr Integrating SHAP analysis with machine learning to predict postpartum hemorrhage in vaginal births
title_full_unstemmed Integrating SHAP analysis with machine learning to predict postpartum hemorrhage in vaginal births
title_short Integrating SHAP analysis with machine learning to predict postpartum hemorrhage in vaginal births
title_sort integrating shap analysis with machine learning to predict postpartum hemorrhage in vaginal births
topic Postpartum hemorrhage
Vaginal births
Interpretable model
XGBoost
SHAP
url https://doi.org/10.1186/s12884-025-07633-w
work_keys_str_mv AT zixuansong integratingshapanalysiswithmachinelearningtopredictpostpartumhemorrhageinvaginalbirths
AT honglin integratingshapanalysiswithmachinelearningtopredictpostpartumhemorrhageinvaginalbirths
AT mengyuanshao integratingshapanalysiswithmachinelearningtopredictpostpartumhemorrhageinvaginalbirths
AT xiaoxuewang integratingshapanalysiswithmachinelearningtopredictpostpartumhemorrhageinvaginalbirths
AT xuetingchen integratingshapanalysiswithmachinelearningtopredictpostpartumhemorrhageinvaginalbirths
AT yangzizhou integratingshapanalysiswithmachinelearningtopredictpostpartumhemorrhageinvaginalbirths
AT dandanzhang integratingshapanalysiswithmachinelearningtopredictpostpartumhemorrhageinvaginalbirths