Systematic review of machine learning approaches for predicting sickle cell crisis and mortality risk at the climate-health nexus.

In 30 seconds

This systematic review analyzed 34 studies on machine learning models predicting sickle cell anemia (SCA) crises and mortality, focusing on climate-related factors. It found that most models primarily used clinical data, with limited climate integration. However, models that included environmental variables showed improved predictive performance, emphasizing the need for climate-informed approaches in SCA management.

Key findings

• Most predictive models relied mainly on clinical and demographic data, lacking climate integration.
• Models incorporating environmental variables demonstrated improved predictive performance.
• There is a significant underrepresentation of high-burden regions, particularly Sub-Saharan Africa, in existing studies.
• The review suggests developing interdisciplinary, climate-aware machine learning frameworks for better SCA management.

Why it matters

Understanding the interplay between climate factors and sickle cell anemia can enhance predictive models, potentially leading to better management strategies and improved patient outcomes in vulnerable populations.