A dangerous pregnancy complication, placenta accreta spectrum (PAS), is becoming increasingly common and a new artificial intelligence (AI) model offers a promising tool for earlier and more accurate detection. PAS occurs when the placenta abnormally attaches to the uterine wall, often following prior cesarean deliveries. This can lead to life-threatening hemorrhage, organ failure, and even maternal death.
Currently, only about half of all PAS cases are diagnosed during pregnancy, leaving many women vulnerable to severe complications. Traditional screening methods, relying on risk factors and ultrasound examinations, can be inconclusive or lead to misdiagnosis. However, research presented at the Society for Maternal-Fetal Medicine (SMFM) Pregnancy Meeting™ suggests a significant step forward in improving early detection.
AI Model Demonstrates High Accuracy in Detecting Placenta Accreta
Researchers from Baylor College of Medicine in Houston, Texas, have developed an AI model capable of analyzing 2D obstetric ultrasound images to identify the presence of PAS. In a retrospective study, the AI was tested on images from to of 113 patients considered at high risk for PAS who delivered at Texas Children’s Hospital. The results were encouraging: the AI accurately detected all confirmed cases of PAS, with no false negatives. There were two false positives, indicating a high degree of specificity.
“Our team is very excited about the potential clinical implications of this model for accurate and timely diagnosis of PAS,” said Dr. Alexandra L. Hammerquist, a maternal-fetal medicine fellow at Baylor College of Medicine. “We are hopeful that its use as a screening tool will help decrease PAS-related maternal morbidity and mortality.”
The Rising Incidence of Placenta Accreta
The increasing prevalence of PAS is a growing concern for maternal health. Data indicates a significant rise in incidence over recent decades. Between and , the rate of placenta accreta in the United States was 1 in 272 deliveries, a substantial increase from 1 in 2,510 deliveries in the . This trend is closely linked to the increasing rate of cesarean deliveries.
Cesarean delivery is a known risk factor for PAS, as it can lead to scarring on the uterine wall, creating an environment where the placenta is more likely to abnormally implant. The MSD Manual notes that prior cesarean section significantly increases the risk of developing the condition. This increase isn’t limited to the United States; a national survey in France indicates that the incidence of PAS is rising globally, driven by both increased cesarean rates and a trend toward pregnancies at older ages.
Challenges in Current Diagnosis and the Potential of AI
The difficulty in accurately diagnosing PAS before delivery poses a significant challenge to patient care. Without a timely diagnosis, healthcare providers may not be adequately prepared to manage the potential complications, such as massive hemorrhage, which can occur during or after delivery. Current screening methods often rely on identifying risk factors and performing ultrasound examinations, but these methods can be limited by inconclusive findings or misinterpretations.
The AI model developed by researchers at Baylor College of Medicine offers a potential solution to these challenges. By analyzing 2D ultrasound images, the AI can provide a more objective and accurate assessment of the risk of PAS. This could allow healthcare providers to proactively plan for delivery, including assembling a specialized team and preparing for potential blood transfusions or surgical interventions.
Looking Ahead: Integrating AI into Clinical Practice
While the results of this study are promising, further research is needed to validate the AI model’s performance in larger and more diverse populations. It’s also important to determine how best to integrate this technology into existing clinical workflows. The researchers emphasize that the AI model is not intended to replace traditional screening methods but rather to serve as an additional tool to improve diagnostic accuracy.
The potential benefits of widespread adoption of this AI technology are significant. By enabling earlier and more accurate detection of PAS, it could lead to a reduction in maternal morbidity and mortality, ultimately improving outcomes for both mothers and babies. In , approximately 21.4% of deliveries in France were performed by cesarean section, highlighting the potential impact of this technology on a substantial proportion of pregnant women.
This AI-based screening breakthrough represents a convergence of clinical need and technological advancement, poised to deliver meaningful benefits for maternal and fetal health worldwide.
