The Google Brain Science team recently announced that they have developed a computer vision system to identify protein crystals in drugs. The team claims that its accuracy rate is about 94%. Protein crystals determine the shape of cells and are useful for discovering drugs that can be targeted to treat various diseases.
Vincent Vanhoucke, chief scientist of the Google Brain Science team, said in a blog post today: "Every protein recognition is performed hundreds of times. Although it is automated in setup and imaging, the protein crystals identified by the visual recognition system are still There are a lot of erroneous possibilities, and human error is easy to occur.†He also said: “The most important thing is that once you miss out on identifying these protein crystals, it will lead to biomedical advancement and improve the status quo of drug use, and Lose valuable opportunities."
In order to train the artificial intelligence model, Google researchers will collaborate with the crystal result recognition company MARCO.
The collaboration between the Google Brain Science team and MARCO has been open sourced, published on GitHub, and detailed in a paper published in the PLOS One journal.
Now we can see a trend, a growing number of large technology companies and record companies to enter the health care field, especially in the provision affecting service.
Baidu Research announced last month that they have created an algorithm that is more accurate and better than a human pathologist to identify breast tumors. Early results also showed that artificial intelligence is more accurate than humans in monitoring skin cancer.
Similarly, last month, Google researchers published the results of a predictive deep learning model in the NPJ Journal of Digital Medicine. The model can determine if a patient needs to be re-admitted for treatment, as well as the length of their stay, and the likelihood of their death during hospitalization.
In the early part of this year, Google shared a computer vision model. This model can monitor the condition of diabetic retinopathy, which is a key indicator of cardiovascular disease.
Verily Life Sciences - like Google, is a subsidiary of Alphabet and interested in the field of drug identification/excavation. In addition, Verily announced on Wednesday that it will work with ResMed to explore sleep-disorders, which may be the cause of weight gain or heart failure.
Cold Chain Products,Medical Refrigerator,Platelet Incubator,Car Refrigerator
Guangdong Widinlsa International Co.Ltd , https://www.gzwidinlsa.com