Synopsis
A new AI model has identified previously unknown risk factors associated with serious pregnancy complications, including stillbirth, highlighting the need for personalized assessments and interventions.Key Takeaways
- A novel AI model detects hidden risk factors for pregnancy complications.
- Research involved analyzing data from nearly 10,000 pregnancies.
- 10-fold variations in risk were observed among infants treated equally.
- Factors like fetal sex and maternal diabetes influence risk levels.
- The model aims for more personalized pregnancy care.
New Delhi, Jan 30 (NationPress) A groundbreaking artificial intelligence model has unveiled previously unrecognized risk factor combinations associated with severe adverse pregnancy outcomes, including stillbirth. A group of researchers from the Universities of Utah and Brown conducted an AI analysis involving nearly 10,000 pregnancies across the nation.
This analysis included various social and physical attributes, ranging from the level of social support for pregnant individuals to their blood pressure, medical histories, and fetal weights, alongside each pregnancy's outcome.
The results, published in the journal BMC Pregnancy and Childbirth, revealed that there could be as much as a 10-fold variation in risk for infants who are currently managed similarly according to clinical protocols.
Factors such as fetal sex, the presence or absence of pre-existing diabetes, and the existence of fetal anomalies like heart defects could play a significant role in determining this risk.
The AI model identified a “truly unexpected” set of factors that indicated elevated risk, according to Nathan Blue from Utah’s Department of Obstetrics and Gynecology.
This model is a step toward enabling “more personalized risk assessment and pregnancy care,” he noted.
Findings indicated that female infants may face a greater risk than males when the mother has pre-existing diabetes, despite the general trend showing female fetuses to be at slightly lower risk for complications.
“The AI model revealed insights that could enhance risk assessment, which even the most skilled clinicians might overlook,” said Blue.
The researchers specifically aimed to refine risk estimates for fetuses in the lowest 10 percent for weight, excluding those in the bottom 3 percent.
These infants are small enough to warrant concern but typically remain healthy.
While existing clinical guidelines recommend vigilant medical monitoring for all such pregnancies, the team discovered that within this fetal weight category, the risk of adverse pregnancy outcomes varied significantly.
