Home > News content

The Baidu AI developer conference shows the parasite identification system

via:快科技     time:2018/7/4 12:41:36     readed:126

Specifically,With the deep learning ability of Baidu AI technology, Dr. Chen has been able to set up a set of parasite identification systems in a few short months to identify more than 10 parasite eggs, several of which are more accurate than 97% in small range clinical trials, comparable to 20 years of experience experts in bed.

As for specific uses,This system can help grass-roots doctors to use microscopes to identify the eggs of the insects, improve the medical ability of the grass-roots doctors, and save the time for the herdsmen to turn hundreds of kilometers to the hospital.

Deep learning is the biggest advantage in the current AI project, including the previous Alfa dogs that beat Li Shishi, which also accumulated a lot more than the human chess player through an uninterrupted game of self play.

What is even more frightening is that the efficiency of AI training is far greater than that of human beings.Recently, Open AI, who defeated Dota amateur, has completed self training for more than 180 years in a short time.

This is why Baidu AI has been able to compete with clinical experts for 20 years in a few months.

In addition, the image recognition of human parasites is also one of the research topics in the medical field. Human parasites are mainly divided into protozoa, worms and arthropods. The most common worms in China are worms, and the main parasite eggs are divided into 10 species.

China IT News APP

Download China IT News APP

Please rate this news

The average score will be displayed after you score.

Post comment

Do not see clearly? Click for a new code.

User comments