Scientists from the University of Geneva (UNIGE) have developed a method for diagnosing colorectal cancer from stool samples using machine learning. The accuracy of this approach reaches 90% - almost as good as colonoscopy (94%), which is currently considered the "gold standard" for detecting the disease.
This is reported by ScienceDaily.
The team has created the first complete catalogue of intestinal bacteria at the subspecies level. This allowed them to track subtle differences in the functions of microorganisms that can both contribute to the development of cancer and be neutral. Using this catalogue together with clinical data, the researchers created a model that can predict the presence of colorectal cancer based only on the microbial composition in stool samples.
The method has the potential to become a simple and inexpensive screening tool for early detection of the disease, especially given the increasing number of cases among young people. A clinical trial is currently being prepared in conjunction with the University Hospitals of Geneva (HUG), which will allow testing at what stages cancer and precancerous changes can be detected using this method.
Scientists believe that the technology can also be used in the diagnosis of other diseases, as the analysis of bacterial subspecies opens up new opportunities for understanding the connection between the microbiota and human health.
As a reminder, almost half of Ukrainians have had their mental health deteriorate over the past three years due to the russian military threat.
Who we are: About us, Contacts. How we write news and our principles: Editorial code. We did our best. If you found this valuable – please support us.
To request a correction, please send an email.