A research team at Purdue University's Weldon School of Biomedical Engineering is taking part in a two-year study to evaluate new ways to monitor the health of pregnant women in Africa. The study aims to inform future efforts to reduce maternal mortality rates.
Young Kim, a professor of biomedical engineering at Purdue, has developed a technology that could help identify women who are at higher risk for preeclampsia. Preeclampsia is a serious pregnancy complication that can lead to maternal death, premature birth, stillbirth, and neonatal death worldwide. The project is funded by the Gates Foundation's Grand Challenge awards.
The innovation uses a patented, noninvasive computer-vision method called mHealth conjunctiva AI imaging. This approach analyzes smartphone photographs of the eyeball to predict early signs of preeclampsia. It works by extracting microvascular patterns from photos of the conjunctiva—the thin membrane covering the inner eyelids and white part of the eye. The technology will be tested in partnership with AMPATH in Kenya.
Kim disclosed these computer- and color-vision methods to the Purdue Innovates Office of Technology Commercialization, which has applied for patents on this intellectual property.
This research is part of Purdue's One Health initiative, which focuses on projects at the intersection of human, animal, and plant health.
Preeclampsia usually develops after the 20th week of pregnancy or during the postpartum period and involves persistent high blood pressure. Early diagnosis and treatment are important for preventing complications for both mother and baby.
According to the World Health Organization, preeclampsia affects up to 8% of pregnancies globally and causes about 46,000 maternal deaths as well as half a million fetal or newborn deaths each year.
The condition often begins without obvious symptoms, making it difficult to diagnose early enough. If left untreated, it can be fatal for both mother and child. Signs include high blood pressure and protein in urine indicating kidney damage or other organ problems.
During this study, researchers will use computer-vision techniques developed at Purdue on smartphone photos collected from 1,600 pregnant women recruited through Moi University via AMPATH Kenya—a group active within local communities.
Kim noted that past studies have shown links between microvascular abnormalities such as narrowing vessels with high blood pressure: "Several previous studies found the alterations preceded the clinical onset of preeclampsia," he said. "Microvascular changes in the retina were observed during the early weeks of gestation and were linked to increased peripheral resistance before blood pressure rose."
He explained that traditional computer-vision analysis requires specialized devices like retinal fundus imaging systems but their mobile solution overcomes this challenge: "Our noninvasive solution eliminates the need for specialized equipment," he said. "Smartphones have recently transformed health care in resource-limited settings where community health workers are often equipped with mobile health apps to connect with health care professionals even from remote areas."
Looking ahead, Kim outlined key next steps: "Our first step will be to refine the prediction model to specifically target preeclampsia rather than general maternal hypertension," he said. "Next, we'll develop a minimally viable mobile app to support scalable validation."
