A team from Dexeus Mujer’s Gynecological Diagnostic Imaging Service (DGI) has identified in a study that the use of artificial intelligence (AI) in the diagnosis of breast cancer would allow a 30% reduction in biopsies in lesions located in the boundary between the cases considered …
A team from Dexeus Mujer’s Gynecological Diagnostic Imaging Service (DGI) has identified in a study that the use of artificial intelligence (AI) in the diagnosis of breast cancer would allow a 30% reduction in biopsies in lesions located in the limit between the cases considered low risk and those considered to be more risky.
The scientific journal ‘Diagnostics’ has published the results of the work in which it has analyzed 403 breast ultrasounds obtained in the gynecological check-ups carried out by the center on women of different ages, it explained in a statement on Thursday.
The diagnosis of breast cancer is based on the study of images obtained through mammography, but in the case of dense or fibrous breasts it is more difficult and it is often necessary to carry out other complementary tests..
According to Dexeus Mujer, of every 100 women who have a mammogram, 70% require an ultrasound to complete the study and, of these, between two and three will have a biopsy, of which only one case will confirm cancer.
The AI-based system used by the center uses technology that has been developed by the North American company Koios Medical.
It is a system that uses algorithms to assess the probability that a lesion is malignant: they have used a database of more than 400,000 images of breast lesions for its development.
The system is authorized by the Food and Drug Administration (FDA) and the European Medicines Agency (EMA), and “one of its advantages is that it can be easily incorporated into conventional ultrasound machines“.
