Friday July 13, 2018

Automating Drug Discoveries Using Computer Vision

Over at the Google AI blog they're talking about computer vision and how they are using it to find protein crystallization. According to Google, computer vision will replace human researchers and reduce human error while finding individual protein crystals that can be targeted by new drugs. This technique currently has an accuracy of 94%. That's pretty accurate and shows how important machine learning really is to science. This is the kind of thing that's going to help us better treat many diseases.

If you want to use your spare computer cycles to help in medical research you should stop by our Distributed Computing Forum and chat with the folks there.

News Image

...we discuss how we used some of the most recent architectures of deep convolutional networks and customized them to achieve an accuracy of more than 94% on the visual recognition task of identifying protein crystals. In order to spur further research in this area, we have made the data freely accessible, and open-sourced our model as part of the TensorFlow research model repository, and available to researchers as a Cloud ML Engine endpoint.