London (IT Boltwise)-The early detection of diabetes risks could be revolutionized thanks to a new AI model. Research Institutes researchers have developed a system that goes far beyond traditional HBA1C tests.
The diagnosis and prevention of diabetes could be significantly improved by the use of artificial intelligence (AI). A new model developed by researchers from the SCRIPPS Research Institute uses data from continuous glucose measurement systems (CGM) and combines them with information about the intestinal microbioma, nutrition, physical activity and genetic factors. This comprehensive data analysis enables early signs of a diabetes risk that may be overlooked in conventional HBA1C tests.
The HBA1C value, which reflects the average blood sugar level of recent months, is an established indicator in diabetes diagnostics. However, this value alone is not sufficient to determine the individual risk of a person for the development of prediabetes or type 2 diabetes. The new AI model shows that two people with identical HBA1C values can have very different risk profiles. By analyzing additional data, such as the duration until the normalization of blood sugar tips, the model can make more precise predictions.
A remarkable finding of the study is the realization that the time the blood sugar level needs to fall again after an increase is a clear indicator of a diabetes risk. In people with type 2 diabetes, this process often takes longer than 100 minutes, while healthier people return to their initial value faster. It was also shown that a diverse intestinal microbioma and a high level of activity correlate with better blood sugar control.
The study, which was carried out completely remote, recruited over 1000 participants from the United States who collected their data over a period of ten days. This innovative approach made it possible to carry out the study without visiting a clinic, which required a new infrastructure for digital clinical studies. The researchers plan to continue to observe the participants in order to check the long -term accuracy of the predictions and to validate the model for wider clinical application.
The results of this research could pave the way for personalized treatment strategies that aim at lifestyle changes and early interventions. Future versions of the model could not only be used by specialist staff in clinics, but also by individuals at home in order to achieve a better understanding and better control over one’s own risk of diabetes.
Enter here for free!
Additions and information to the editorial team by email to DE-Info[at]it-boltwise.de. Since we cannot rule out rarely occurring AI Hallucinations in the event of AI generated news and content, we ask you to contact and inform us for false information and misinformation via email. Please do not forget to name the article headline in the email: “AI model for the early detection of diabetes risks”.
