Belarus’ NAS revealed how neural networks help in medicine
Neural networks began to penetrate into all spheres of human life recently. Although now there are disputes about the use of AI, there are still a huge number of examples when its use benefit people. One such area is medicine. Eduard Snezhko, who heads the biomedical image analysis laboratory of the United Institute for Informatics Problems at the National Academy of Sciences of Belarus, spoke about how AI helps people work in the field of image processing during his talk with STV TV channel.
According to Eduard Snezhko, since 2014 there has been a new wave of interest in medicine in the use of neural networks for processing volumetric data.
“Our laboratory has been processing images, in particular medical ones, for a long time,” Snezhko explained. “Recently, probably since 2014, a new wave of interest in deep learning methods has begun, particularly in machine learning methods, which enable us to process huge amounts of data that have accumulated in our field. This applies to X-ray, MRI, and CT images, etc.”
Eduard Snezhko said that AI helps specialists to see some of the most important aspects in a large amount of data, “We have a human screening programme operating in Belarus. That is, once a year, each person must undergo an examination and receive a two-dimensional image of their chest, so that doctors can detect any pathological processes at an early stage. This is a very large amount of data, and it is natural that specialists may miss some important detail when working with them. During processing AI helps to draw doctor’s attention to some pathologies in the patient.”
Eduard Snezhko also noted that neural networks are a machine that a person uses. He compared them to a mechanical arm that amplifies and speeds up the work of specialists with a lot of data. But in the end, the decision is still made by the medical professional.
