Approximately 10% of infants born in the U.S. in 2021 were delivered prematurely, before 37 weeks of pregnancy, according to the Centers for Disease Control and Prevention (CDC). Preterm births account for about 16% of infant deaths. Researchers from Washington University in St. Louis, Missouri, have developed an artificial intelligence (AI) model that uses deep learning to predict preterm births by analyzing electrical activity in a woman’s uterus during pregnancy. The model was tested in a study published in the medical journal PLOS One.
The researchers were surprised to find that the AI model could accurately predict preterm birth as early as the 31st week of pregnancy, up to the 37th week. The model automatically learned the most relevant features from the data related to preterm birth. Additionally, the findings suggest that preterm birth is not simply an early end to a pregnancy, but rather an abnormal physiological condition.
During the study, the researchers recorded electrical activity in the uterus using electrohysterograms (EHGs) and combined the data with other medical information. Nearly 19% of the women in the study delivered prematurely. The advantage of this approach is that it is cost-effective and can be used with shorter EHG recordings, making it easier to implement in clinical or even home settings.
The researchers believe that this method should be incorporated into regular pregnancy check-ups and adopted by hospitals and obstetricians to enable pregnant women to take necessary precautions and seek appropriate care. They suggest creating a dedicated device to implement their method.
While the AI model shows promise in predicting preterm birth, further studies with larger patient populations are needed before it can be widely adopted. The use of EHG measurements as a standard evaluation tool would require hospitals and offices to invest in new equipment. False positives and the need for improved treatments for preterm labor are also considerations.
The study’s main limitations include the relatively small number of patients involved and the difficulty in interpreting how the algorithm makes predictions. Despite these challenges, machine learning algorithms like this one have the potential to improve healthcare, but additional research is needed to fully understand how to integrate them into clinical practice.
Denial of responsibility! VigourTimes is an automatic aggregator of Global media. In each content, the hyperlink to the primary source is specified. All trademarks belong to their rightful owners, and all materials to their authors. For any complaint, please reach us at – [email protected]. We will take necessary action within 24 hours.
