BLOOM · Birth and Labor Outcome Optimization through Multimodal AI

Cardiotocography (CTG) is a vital tool for monitoring fetal well-being during labor, recording fetal heart rate and uterine contractions to detect issues like hypoxia, which can cause neonatal deaths or long-term neurological damage. Early detection through CTG enables timely interventions, such as emergency cesarean sections, reducing risks. However, interpreting CTG data is challenging and depends heavily on clinician expertise, making assessments sometimes subjective and prone to error. This has driven interest in integrating artificial intelligence (AI) into CTG analysis to support clinicians with more accurate and timely decision-making. While AI shows promise in enhancing CTG interpretation, a significant challenge remains the absence of a universally accepted standard for assessing fetal outcomes. Many existing methods rely on surrogate markers or rigid scoring systems that fail to account for the complex physiological responses of individual fetuses. A pathophysiological approach, which considers unique fetal compensatory mechanisms and maternal health factors like hypertension or gestational diabetes, can provide a more nuanced and precise assessment of fetal well-being. The Birth and Labor Outcome Optimization through Multimodal AI (BLOOM) project is designed to address these challenges. It aims to develop, implement, and validate an advanced AI system specifically for use during labor. This system will integrate multimodal data, including fetal growth patterns from ultrasound screenings, maternal health conditions from electronic health records (EHR), and pathophysiological interpretations of CTG data. By combining these personalized factors with cutting-edge AI technologies, BLOOM seeks to create a more accurate and effective tool for assessing fetal well-being during labor, ultimately improving clinical outcomes and personalizing care.