The system reviews images of patients and diagnoses cancer. She acts like a doctor

The authors studied the mechanics of the human brain and its ability to distinguish different parts of an image. On the

Based on this information, they created a new system that replicates the features of human vision.

The new system can be used in differentfields: in robotics, multimedia communications, video surveillance, automated image editing, and even in the search for tumors in medical images.

Research Group on MultimediaComputing at Cardiff University is currently planning to test a system to help radiologists find lesions in medical images. The goal is to increase the speed, accuracy and sensitivity of medical diagnostics.

In their study, the team useda deep learning computer algorithm called a convolutional neural network. They mimic the work of neurons in the human brain, in particular for the visual cortex.

This type of algorithm is ideal for image processing. It can assign different degrees of importance to different objects or aspects within the image itself.

The team used a huge database of images, where all the photos have already been studied by people and the most interesting areas have been marked.

The next step, the team says, is working with radiologists. They want to understand how their AI can help doctors find important areas in medical images.

Read more:

Astronomers have calculated when the planets will change their orbits and chaos will come

The space probe flew 200 km from Mercury. Look what he saw

The unique spaceplane will deliver cargo to the ISS. He is not like the others