I-Corps: Translation potential of computer modeling and machine learning for analyzing Doppler ultrasound data to predict placental insufficiency

This I-Corps project is based on the development of artificial intelligence (AI) technology for predicting placental insufficiency from ultrasound images. Placental insufficiency is a condition where the placenta fails to deliver an adequate supply of nutrients and oxygen to the fetus. This condition impacts 6 million pregnancies in the United States. Current diagnostic methods for placental insufficiency are reactive, relying on symptom presentation and/or emergencies and this limits the ability of doctors to prevent adverse outcomes and fetal death. The goal of this technology is to accurately predict the likelihood of placental insufficiency. A key marker in determining placental insufficiency comes from understanding the role of blood flow in placental health. By examining blood flow and identifying early signs of insufficiency, clinicians may identify at-risk mothers and provide appropriate planning and intervention throughout the pregnancy. In addition, this technology may be used in both urban and rural settings to improve outcomes in both maternal and fetal healthcare.

This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of artificial intelligence (AI)-based, predictive technology to examine the maternal-fetal interface by analyzing Doppler ultrasound images. Placental insufficiency, which often goes undetected until it starts to impact fetal growth, is limited in terms of early predictability. This technology is based on detailed digital twins, numerical modeling, and machine learning and may be used to determine blood flow patterns in the placenta. This solution differs from existing AI technologies by looking at multiple parameters (e.g., blood flow, vessel geometry, patient history, and socioeconomic factors) to generate a score that identifies at-risk mothers during the first and second trimesters. The goal is to provide clinicians with a new decision support system to identify and engage at-risk mothers to improve pregnancy outcomes. This AI predictive technology has the potential to change the management of placental problems by offering earlier detection, greater accuracy, and tailored intervention strategies. This solution may help physicians avoid serious complications such as fetal growth restriction, stillbirth, or pre-eclampsia, ultimately reducing maternal and fetal morbidity and mortality and improving patient outcomes.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Award Number: 2524269
Principal Investigator: Sunghan Kim
Funds Obligated: $50,000
State: NC