Artificial Intelligence To Predict Pre-Term Birth

Researchers from across the world ventured to find a technological solution to what poses significant risks and is a major public health concern to pregnant women, preterm birth. It affects approximately 10% of pregnancies globally.

Babies born preterm succumb to illness and disorders like long-term disability such as neurological disorders, behavioral problems, developmental delays, and mental conditions.

To tackle this problem, a team of researchers found ways to integrate AI to offer novel approaches to predict, diagnose, detect, and monitor perinatal health. This is made possible by the use of machine learning, a commonly used artificial intelligence method to analyze electrical activity in the uterus and thereby accurately predict birth defects. The efficiency of this method has been validated.

According to a report by WHO and UNICEF, approximately 13.4 million babies were born preterm in 2020 with nearly 1 million succumbing to complications leading to infant mortality. Research also suggests that a majority of preterm births occur in South Asia and sub-Saharan Africa but it remains to be a pressing issue globally.

Through the exploration of innovative interventions, researchers from Washington University in St. Louis, employed AI to analyze electrical activity in the uterus. This study was approved and published in the medical journal PLOS One.

During the study, the researchers performed electrohysterograms (EHGs), which use electrodes on the abdomen to record electrical activity in the uterus. They took recordings of these electrical currents from 159 pregnant women who were at least 26 weeks along and “trained” the AI model on that data.

Dr. Arye Nehorai, PhD, Professor of electrical engineering at Washington, discovered that it is possible to take data as possible as the 31st week and predict preterm birth up to the 37th week. This finding suggests that preterm birth is not merely an early delivery but rather an abnormal physiological condition.

The researchers recommend the integration of this AI-based method into routine pregnancy check-ups, encouraging hospitals and obstetricians to adopt it. By incorporating this approach, pregnant women can receive appropriate care and make necessary lifestyle adjustments to safeguard the well-being of their babies. Another advantage is that the method is cost-effective, making it accessible to hospitals worldwide. However, uncertainty remains as to when the method will be adopted worldwide by hospitals

According to the finding of the report, Born Too Soon: Decade of Action, it highlights that Kenya is among the countries that  have recorded moderately high preterm births at below 10 percent translating to 127,500 preterm babies in 2020.

This means that the total number of preterm babies that were born in Kenya in 2020 was 127,500 out of 1,445,900 preterm babies born in that year.

However, the researchers note that uncertainty remains regarding the timeline for the widespread availability of this test. Despite this obstacle, the potential impact of implementing this technology is significant.