AUTHOR=Shah Santosh Yogendra , Saxena Sumant , Rani Satya Pavitra , Nelaturi Naresh , Gill Sheena , Tippett Barr Beth , Were Joyce , Khagayi Sammy , Ouma Gregory , Akelo Victor , Norwitz Errol R. , Ramakrishnan Rama , Onyango Dickens , Teltumbade Manoj 

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

JOURNAL=Frontiers in Global Women's Health

VOLUME=Volume 4 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/global-womens-health/articles/10.3389/fgwh.2023.1161157

DOI=10.3389/fgwh.2023.1161157

ISSN=2673-5059

ABSTRACT=Postpartum hemorrhage (PPH) is a major cause of maternal mortality worldwide and in low-and middle-income countries. Antenatal, intrapartum, and postnatal visit data from Kenya Antenatal and Postnatal Care Research Collective cohort were used to predict postpartum hemorrhage. Four machine learning modelslogistic regression, naïve Bayes, decision tree, and random forestwere 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. The 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. 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.