Reducing Neonatal and Infant Mortality with a Birth Oximetry Routine for Newborns in Pakistan

Pakistan has one of the highest neonatal mortality worldwide with the main reported causes being sepsis and congenital heart defects. Despite many efforts to address these, the mortality rate has reached a plateau. There is thus a need to institute innovative technology such as pulse oximetry with machine learning to screen for neonatal wellness. 

Machine learning will develop an automated decision making algorithm on the pulse oximetry data for community health workers to facilitate early diagnosis and appropriate referral for at-risk newborns.

Objectives: Our aim was assessing the value of pulse oximeter and WHO Young Infant Clinical signs independently, in combination, and with machine learning (ML) from clinical features, to detect at risk neonate in a low- and middle-income country.  

Team: Zahra Hoodbhoy (AKU), Babar Hasan (SIUT), Devyani Chowdhury (Cardiology Care for Children USA) and Bart Bijnens (UPF)

Site: Ibrahim Hyderi, Karachi

Timeline: Project completed

Sponsor: Islamic Development Bank

Dissemination, presentations:  

Poster presented at Grand Challenges in Belgium in 2022
POx poster-AKU Paediatrics and Child Health.pdf

Hasan BS, Hoodbhoy Z, Khan A, Nogueira M, Bijnens B, Chowdhury D. Can machine learning methods be used for identification of at-risk neonates in low-resource settings? A prospective cohort study. BMJ Paediatrics Open. 2023;7(1).​