Chinese research suggests that the risk of miscarriage can be detected before pregnancy with the help of an AI model. The team, led by Dr. Li Wei, a professor in the Chemistry Department at Peking University, has developed the AI model using a machine learning algorithm called XGBoost.
According to research, XGBoost can predict the likelihood of miscarriage prior to pregnancy by analyzing the level of serum metabolites, especially histidine. This research has been published in the Journal of Reproductive Medicine. It claims that if the level of certain metabolites such as histidine is high, the risk of miscarriage may increase.
The report states that women who have faced miscarriage more than twice are more likely to suffer from this as compared to normal women. However, there are ethical challenges with this AI model. Researchers said that data security is the biggest challenge on such platforms. In addition, it is important to remember that women should be free to take decisions related to her health and freely share correct information related to her health without any pressure.
The prototype of this model has been built in China for global use and further testing, but the researchers say that there are plans to expand it to many more countries in the future.
1. New uses of AI: Scientists have created an AI-based model that can identify the risk of miscarriage in women before pregnancy with up to 87% accuracy..
2. Importance of serum metabolites: Studies indicate that serum metabolites, especially ‘histidine’, present in a woman’s body before pregnancy may cause miscarriage.
3. Interactive Platform: Scientists have also created a digital platform where women can share their health details to predict miscarriage.
