AI helps pathologists find signs of preeclampsia, advancing diagnosis and treatment
Preeclampsia is one of the leading causes of pregnancy-related death. Even after a healthy delivery, mother and baby can go home only to show signs of a postpartum hypertensive disorder days or weeks later.
Preeclampsia, a leading cause of pregnancy-related deaths, can manifest in mothers and babies weeks after delivery, even after a healthy delivery. Pathologists study placental blood vessels to identify disease signs, but a shortage of trained personnel means less than 20% of placentas in the U.S. are screened. Carnegie Mellon University and UPMC researchers developed a machine learning algorithm to expedite this process.
The algorithm identifies decidual vasculopathy, a disease of maternal blood vessels in the placenta associated with postpartum preeclampsia, by analyzing the spatial organization of extravillous trophoblast cells and red blood cells within images of placental vessels. By calculating a morphology separation score, the model distinguishes healthy from diseased vessels and explains why it reached that conclusion, offering a biologically relevant and interpretable measure.
This explainable AI model could help expand screening for preeclampsia and potentially improve the diagnosis of other diseases with early biomarker indicators.
Written by urgent.news from Medical Xpress's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
