Prediction of postpartum hemorrhage (PPH) using machine learning algorithms in a Kenyan population

Frontiers in Global Women's Health

Global Women's Health

Joyce Were¹  Sammy Khagayi¹  Gregory Ouma¹  Victor Akelo²  Manoj Teltumbade³  Sheena Gill³  Naresh Nelaturi³  Sumant Saxena³  Satya Pavitra Rani³  Santosh Yogendra Shah³  Errol R. Norwitz⁴  Dickens Onyango⁵  Beth Tippett Barr⁶  Rama Ramakrishnan⁷ 

[1] Center for Global Health Research, Kenya Medical Research Institute, Kisumu, Kenya;Center for Global Health, U.S. Centers for Disease Control and Prevention, Kisumu, Kenya;CognitiveCare Inc., Milpitas, CA, United States;Department of Obstetrics and Gynecology, Tufts University School of Medicine, Boston, MA, United States;Kisumu County Department of Health, Kisumu, Kenya;Office of the Director, Nyanja Health Research Institute, Salima, Malawi;Operations Research and Statistics, MIT Sloan School of Management, Cambridge, MA, United States;

关键词: maternal health;    machine learning;    pregnancy;    postpartum hemorrhage;    risk prediction;    LMICs;   

DOI  :  10.3389/fgwh.2023.1161157

 received in 2023-02-08, accepted in 2023-07-10,  发布年份 2023

来源: Frontiers

【 摘 要 】

IntroductionPostpartum hemorrhage (PPH) is a significant cause of maternal mortality worldwide, particularly in low- and middle-income countries. It is essential to develop effective prediction models to identify women at risk of PPH and implement appropriate interventions to reduce maternal morbidity and mortality. This study aims to predict the occurrence of postpartum hemorrhage using machine learning models based on antenatal, intrapartum, and postnatal visit data obtained from the Kenya Antenatal and Postnatal Care Research Collective cohort.MethodFour machine learning models – logistic regression, naïve Bayes, decision tree, and random forest – were constructed using 67% training data (1,056/1,576). The training data was further split into 67% for model building and 33% cross validation. Once the models are built, the remaining 33% (520/1,576) independent test data was used for external validation to confirm the models' performance. Models were fine-tuned using feature selection through extra tree classifier technique. Model performance was assessed using accuracy, sensitivity, and area under the curve (AUC) of the receiver operating characteristics (ROC) curve.ResultThe naïve Bayes model performed best with 0.95 accuracy, 0.97 specificity, and 0.76 AUC. Seven factors (anemia, limited prenatal care, hemoglobin concentrations, signs of pallor at intrapartum, intrapartum systolic blood pressure, intrapartum diastolic blood pressure, and intrapartum respiratory rate) were associated with PPH prediction in Kenyan population.DiscussionThis study demonstrates the potential of machine learning models in predicting PPH in the Kenyan population. Future studies with larger datasets and more PPH cases should be conducted to improve prediction performance of machine learning model. Such prediction algorithms would immensely help to construct a personalized obstetric path for each pregnant patient, improve resource allocation, and reduce maternal mortality and morbidity.

【 授权许可】

Unknown   
© 2023 Shah, Saxena, Rani, Nelaturi, Gill, Tippett Barr, Were, Khagayi, Ouma, Akelo, Norwitz, Ramakrishnan, Onyango and Teltumbade.