Quick Buzz Feed
A Cambodian physician argues that artificial intelligence can help improve maternal health outcomes, but only if accompanied by structural reforms
This section highlights the potential of AI tools, such as portable ultrasounds and early-warning systems, in reducing maternal mortality in low- and middle-income countries. It emphasizes that while AI demonstrates significant statistical improvements, its practical effectiveness hinges on adequate workforce training, integrated clinical workflows, and institutional preparedness for emergencies. The author, reflecting on a 'Design for Implementation' conference, underscores that clinician engagement and a clear understanding of AI tools are paramount. AI should augment, not replace, human clinical judgment. Successful applications include early detection of postpartum hemorrhage and preeclampsia, and digital health interventions like the QUALMAT project in Ghana and Tanzania, which improved antenatal and intrapartum care by guiding health workers through recommended steps in real time.
This section addresses the critical need for robust infrastructure, trained staff, and effective governance for AI tools to function sustainably in low-resource maternity wards. Drawing on Cambodia's experience, where increased access to facility deliveries did not lead to corresponding improvements in managing complications, the author argues that many AI pilots fail because they do not account for real-world health system challenges. Sustainable implementation requires reliable infrastructure, clinical staff trained to interpret and act on AI outputs, and clear institutional governance structures. The author urges Cambodia's Ministry of Health and hospital leaders to prioritize governance reforms. Donors are also encouraged to tie AI funding to demonstrated governance readiness and to provide support for implementation science to ensure technology effectively reaches the bedside.