Imaging Software Simulates Childbirth

CHICAGO - Great strides have been made in the past century in ensuring women's safety during childbirth. Now, new imaging software presented Tuesday at RSNA furthers this by simulating childbirth to predict difficult deliveries.

Because of the upright position of humans and the large head size of a fetus, the movement of a fetus through the birth canal of a woman is more complicated than that of other animals, leading to a higher risk of dystocia, or difficult labor. Olivier Ami, MD, of Antoine Béclères Hospital at Université Paris Sud, France, and colleagues used software called Predibirth in conjunction with MR images to study 24 pregnant women and their delivery outcomes.

They produced 3-D reconstructions of the fetus and each woman's pelvic measurements, with which they simulated 72 possible trajectories of the fetal head through the birth canal. Animation of the fetus was obtained by the adjunction of a skeleton to the 3-D reconstruction of the fetus.

“The computer simulates what we call collision detection,” Ami said. The software tries to find a trajectory of the given fetal head through the pelvis, and a collision is detected if it is not possible. Based on these results, each mother was scored on her likelihood to have a favorable childbirth.

In a highly favorable birth, there is no potential collision detected. In a mildly favorable birth, there is less than 3 percent of fetal head volume reduction, as the head may be slightly molded during childbirth. If the fetal head volume reduction is over 3 percent, there is an unfavorable birth, which corresponds to potential brain damage or inability to deliver vaginally.

The Predibirth scores in the study were computed retrospectively and compared with actual delivery outcomes. Of the 24 pregnant women, the 13 who delivered vaginally were all scored as highly favorable by the simulator. Three women who had elective cesarean sections (C-section) also had scores for high dystocia risk.

Emergency C-section for obstructed labor occurred in three patients, who were estimated to have high risk for dystocia. Three instrumental extractions, meaning use of a vacuum to aid in delivery, were scored as mildly favorable.

“We could've targeted the emergency C-section for obstructed labor, and the instrumental extractions, for planned Cesarean sections,” Ami said.

The imaging tests were performed during the last month of pregnancy. “The more you are performing it close to the delivery, the more it is accurate,” Ami said. Extrapolation curves for the growth of the head are also available when a test is performed earlier in the pregnancy.

Approximately a third of all births in the U.S. are delivered through C-sections. An emergency C-section has significantly higher risk for mortality as compared to a planned C-section.