Revolutionising Obstetric Care with AI: The Mama Track Mobile Partograph

Maternal and fetal health monitoring is critical in ensuring safe labour and delivery. Hence, the World Health Organization (WHO) is recommending the use of partographs for labour monitoring. Thus, various computer-aided diagnosis (CAD) systems have recently been developed and play a vital role in supporting obstetricians to evaluate the fetal state. Despite such advancement in the automation of CAD systems. However, prediction of fetal conditions, including prolonged labour, obstructed labour, postpartum haemorrhage, and fetal distress, has been a challenge due to not only visual misinterpretation of the recorded values but also gaps in training. This results in a significant inter-observer variability, which causes delayed interventions and increased maternal-fetal complications. In an attempt to address the challenge, this study has proposed an AI-based mobile partograph (Mama Track Mobile Partograph) that relies on deep learning models designed to assist obstetricians with real-time decision support and predictive analytics. The proposed approach has been evaluated using open-access datasets (CTU-UHB Intrapartum Cardiotocography Database), and the findings suggest that the proposed Mama Track Mobile Partograph has achieved a test set G‑mean of 0.6037, balanced sensitivity (~0.54) and specificity (~0.67), and overall accuracy (0.6257), highlighting its potential to mitigate inter-observer variability in traditional partograph interpretation. Therefore, Mama Track Mobile Partograph is not only a feasible AI solution but also an innovation that can eliminate factors leading to maternal and neonatal morbidity and mortality