Mayo Clinic researchers have found that using artificial intelligence (AI) algorithms to analyze patterns of change in women in labor can help determine whether a successful vaginal birth will occur with good outcomes for mother and baby. The findings were published in PLOS ONE.
This is the first step towards using algorithms to provide powerful guidance to doctors and midwives when making critical decisions during the labor process. Once validated by further research, we believe the algorithm will work in real time, meaning that each new data entry during a pregnant woman’s labor automatically recalculates the risk of an adverse outcome. This can help reduce the rate of caesarean delivery and maternal and neonatal complications.”
Abimbola Famuyide, MD, Mayo Clinic OB-GYN and lead study author
Women in labor understand the importance of periodic cervical exams to assess labor progress. This is an essential step, as it helps obstetricians predict the likelihood of a vaginal birth within a specified period of time. The problem is that cervical dilation during labor varies from person to person, and many important factors can determine the course of labor.
In the study, researchers used data from the National Institute of Child Health and Human Development Eunice Kennedy Shriver Multicenter Occupational Safety Consortium database to create the prediction model. They examined more than 700 clinical and obstetrical factors in 66,586 deliveries from the time of admission and as labor progressed.
The risk prediction model consisted of data known at the time of labor admission, including baseline patient characteristics, the patient’s most recent clinical assessment, and the cumulative progress of labor since admission. ‘admission. The researchers explain that the models can provide an alternative to conventional workflow charts and help individualize clinical decisions using each patient’s baseline and workflow characteristics.
“It’s very individualized for the person in labor,” says Dr. Famuyide. He adds that this will be a powerful tool for remote midwives and doctors, as it will allow time for patients to be transferred from rural or remote settings to the appropriate level of care.
“The ability of the AI algorithm to predict individualized risks during the labor process will not only help reduce adverse birth outcomes, but it may also reduce healthcare costs associated with maternal morbidity in the United States. , which have been estimated at more than 30 billion dollars,” adds Bijan Borah. , Ph.D., Robert D. and Patricia E. Kern Scientific Director of Health Services and Outcomes Research.
Validation studies are underway to assess the results of these models after their implementation in the work units.
This study was conducted in collaboration with scientists from Mayo Clinic Robert D. and the Patricia E. Kern Center for the Science of Health Care Delivery. The authors have declared no competing or potential conflicts of interest.
Shazly, South Africa, et al. (2022) Impact of Labor Characteristics on Maternal and Newborn Labor Outcomes: A Machine Learning Model. PLOS ONE. doi.org/10.1371/journal.pone.0273178.