Dear MEL Topic Readers,
Google's DeepMind AI can accurately detect 50 types of eye disease just by looking at scans
DeepMind seems to like black and white, whether on a go board or in person’s eyes.
In May 2017, DeepMind’s proud AI go playing algorithm AlphaGo beat Ke Jie, the world No.1 ranked player at the time, in a three-game match. Only six month-or-so later, AlphaGo Zero, a version created without using data from human games, surpassed any preceding AlphaGo versions only by playing games against itself. This means DeepMind’s AI algorithms can learn and solve certain problems better than collective and cumulative human ingenuity.
Now, the Google’s AI arm has announced that its algorithm can identify about 50 eye diseases by looking at retinal eye scans almost as well as expert clinicians do. Furthermore, it is capable of making recommendations for treatment. This means that the algorithm could help eye doctors diagnose patients and provide appropriate cares to them much faster than now because the number of experienced eye doctors is limited.
It seems that DeepMind is stepping out of competition against humans and heading for cooperation with humans.
Enjoy reading and learn what AI could do now and in the future.
No comments:
Post a Comment