Traditionally, doctors and midwives have used a technique called pelvimetry to measure the pelvis and try to determine its adequacy for giving birth. But pelvis size is just one factor in how smoothly labor will go, rendering the method largely insufficient.
Scientists in France have been working to take some of the guesswork out of labor predictions. Today, at the Radiological Society of North America's annual meeting, they are presenting results of a study showing that their newly developed software, called Predibirth, predicts birth outcomes quite accurately.
The researchers used their software to process magnetic resonance images of 24 pregnant women, capturing the pelvis and fetus, and then simulating 72 possible trajectories the baby's head might take through the birth canal. The program then uses this data to score the mother's chances of having a normal (vaginal) birth.
"The mechanics of the human birth canal make for a very complicated delivery process, compared to other mammals," Olivier Ami, an obstetrician in the Department of Radiology at Universite Paris Sud, said in a statement. "We now have computer-simulated childbirth to identify potential problems."
Of the 24 women in the study, the 13 who delivered normally all had highly favorable birth outcome scores. Three women who had high-risk scores underwent elective C-sections. Of the five women who underwent emergency C-section, the three with obstructed labor had high-risk scores, and the two who experienced heart rhythm abnormalities had mildly favorable or favorable scores.
"The results in predicting dystocia were highly accurate," Dr. Ami said. "Our simulation predictions seem to be a significant improvement over pelvimetry."
More accurate measurements of labor risks might not only keep C-section rates lower and help identify necessary C-sections before they become emergencies, but these measurements could also better inform those who want to deliver at home whether it is safe to do so.