The team of Dr. Eran Segal, of the Weizmann Institute of Science in Israel, has developed an algorithm to predict the risk of gestational diabetes. Before the patient gets pregnant. The test aims to identify women most at risk, so as to help them change their lifestyle to avoid the onset of the problem.
The researchers analyzed data from over 600,000 gestations. They obtained them from the database of the Israeli organization Clalit Health Services, the most important in the country with a health theme. The goal was to find a system to measure the risk of getting sick before pregnancy, a problem far from trivial.
Gestational diabetes occurs in 3-9% of gestations and represents a risk for mother and child. Just like in other types of diabetes, high blood sugar levels put mother and baby health at risk. Despite the similarity with traditional diabetes, however, gestational diabetes is often unrelated to the presence of cases in the family or in previous gestations. All this makes it much more difficult to predict whether a woman will fall ill or not.
Scientists analyzed data from thousands of women using an artificial intelligence system. Later, they used the algorithm to process more than 2,000 evaluation parameters. These included the results of glucose tolerance tests, blood tests, family history. The system revealed that 9 parameters are enough to predict whether a woman will fall ill.
Thanks to the discovery, few questions and tests will be enough to accurately calculate the risk of disease.