Just as cars have check engine lights, wearable sensors, combined wiht data
analysis, could provide people with early warnings about their health.

According to Shravan Aras, PhD, assistant director of sensor
analysis and smart health platforms at the University of Arizona Health
Sciences Center for Biomedical Informatics and Biostatistics, fitness
trackers offer “an amazing window into our biology and how we operate.”

Aras aims to integrate wearable sensors into research,
optimizing data exploration and analysis. The rise of artificial
intelligence and machine learning offers unprecedented opportunities in this
area.

Aras, who holds a doctorate in computer science, values
collaboration, stating, “I’ve always looked at computer science not in
isolation, but as a collaborative tool – computer science being applied to
different domains to solve really complicated and challenging problems.”

Predicting Labour Onset with AI

one important challenge is predicting the start of labor in pregnant women.
While due dates are typically calculated 40 weeks from the last menstrual
period, actual gestation can vary. Current clinical methods often fall short,
leading to reliance on self-reporting, which can be inaccurate.

Unexpected labor can lead to complications, such as unplanned home births or
insufficient time for intervention in preterm cases.

Elise Erickson, PhD, a certified nurse-midwife and associate
professor of physiology, sought a solution. She collaborated with
Aras on a study using temperature data to predict labor
onset, a common practice in animal studies.

“Most of us wear health and fitness trackers on our wrists or fingers, and
that’s an amazing window into our biology and how we operate.”

Aras explained that while fertility tracking uses daily
temperature readings to determine ovulation, pregnancy involves more complex
hormonal changes. “For labor prediction, daily temperature readings do not
give you a cohesive pattern of when somebody might go into labor.”