Integrating a Machine Learning Algorithm with Solid-State Epidermal Biomarker Sensing to Predict Preeclampsia in Early Pregnancy

Ling-Jun Li of the National University of Singapore in Singapore will develop a diagnostic platform for prediction of preeclampsia early in pregnancy for women at high-risk of developing the condition. Two, existing, Singapore-based pregnancy cohorts will be used to develop and validate an AI-based risk prediction model for hypertensive disorders of pregnancy including preeclampsia. A non-invasive, preeclampsia screening test will then be developed, combining retinal imaging to identify vasculature features plus epidermal detection of the preeclampsia biomarker proteins PIGF and PAPP-A using a wearable forearm patch with a solid-state sensing system. The feasibility and acceptability of this test will be assessed with 60 pregnant women recruited at National University Hospital in Singapore, integrating the clinical data with the AI-based risk prediction model to guide further evaluation of the diagnostic platform.