New brain image labeling system can process 100 thousand images in 30 minutes

Researchers from King's College London's School of Biomedical Engineering and Imaging

automated labeling of MRI imagesbrain, necessary to improve machine learning models. They then recognize images by extracting important labels from radiology reports and accurately assigning them to the corresponding MRI exams. Now more than 100 thousand pictures can be taken in half an hour.

Deep learning usually requires tens of thousandstagged images for best performance in pattern recognition. This is the weakest part in the development of deep learning systems for complex imaging datasets, in particular MRI, which is fundamental for detecting neurological diseases.

“This model made the tasks much easierimage recognition using deep learning, and this will almost certainly accelerate the advent of automated brain MRI readers in the clinic. The potential for patient benefit is enormous, ”the researchers noted.

Scientists have presented a microscope that allows you to see the smallest cellular structures

The study authors note that at least onethe barrier to rapid research has already been overcome, but further problems still need to be addressed. Now scientists want to make their method work in most hospitals that use different scanners.

Previously, scientists from the University of CaliforniaIn Los Angeles, artificial intelligence (AI) was used to identify three new subtypes of multiple sclerosis. Researchers say their findings will help identify those people who are more likely to have the disease progress and help predict treatment more effectively.

Read more

Explained how the universe is reflected near black holes

Solved 180-year-old moth mystery: it was opened six times and named by different names

Media: The Ministry of Digital Affairs accidentally revealed the real scale of the COVID-19 epidemic in Russia