Survey Gathers Clinician Views on Fetal Monitoring AI

AI-powered algorithms are being developed to enhance the interpretation of cardiotocography (CTG), which monitors fetal heart rate and uterine contractions to assess fetal well-being during labor. These systems aim to improve upon the high inter- and intra-observer variability that currently exists in CTG interpretation, a factor that can impact perinatal outcomes. The goal is to create a more objective and consistent analysis to support clinical decision-making. Researchers are employing deep learning and machine learning models to analyze complex CTG data, identifying subtle patterns that may indicate fetal distress. Some studies have reported high accuracy rates, with one AI-driven framework demonstrating a 97.91% accuracy in CTG classification. This technology could be particularly valuable in healthcare settings where immediate expert consultation is not available. The push for AI in fetal monitoring comes as a response to the limitations of current methods, which can have high false-positive rates for detecting fetal distress, leading to unnecessary interventions. Hypoxia during birth is a significant cause of conditions like cerebral palsy, and improved detection is a key goal. Globally, intrapartum-related events account for nearly a quarter of neonatal deaths. Building clinician trust is a critical factor for the successful adoption of these AI tools. Research indicates that transparency, explainability, and usability are key factors that influence a healthcare professional's confidence in medical AI. A 2024 study found that 73% of physicians were reluctant to adopt AI tools due to concerns about their reliability and transparency. The survey on clinician attitudes is part of a broader effort to understand the human factors involved in integrating AI into maternal care. Studies have shown that while clinicians may have positive overall sentiments, they also express concerns about whether AI-assisted decisions align with their own clinical reasoning. The University of Warwick, where some of the researchers involved in the survey are based, has been active in applying machine learning to maternal-fetal health. Previous research from the university developed a test using machine learning to accurately predict the risk of premature birth by analyzing vaginal swabs. Beyond CTG interpretation, AI is being explored for various applications in prenatal care, including analyzing ultrasound images, predicting pregnancy complications like preeclampsia, and personalizing risk assessments. These technologies have the potential to automate repetitive tasks, allowing clinicians to focus more on direct patient care. However, the integration of AI in fetal medicine is not without challenges. Ethical considerations around data privacy, the potential for algorithmic bias, and the need for extensive validation in diverse populations are significant hurdles that need to be addressed.