It seems that Google has been betting its tools of artificial intelligence in the medical area, since previously, the Google has guaranteed that blindness caused by diabetes can be diagnosed in advance, thus giving the possibility of being avoided.
Making the decision and delimiting whether a patient has cancer or not usually involves meticulously trained professionals, in which the disease can be detected by doctors through tissue samples over weeks and months. But an artificial intelligence (AI) program from A, you may be able to do it much faster.
According to the document, Google is working hard to be able to tell the difference between healthy and cancerous tissue, as well as find out if metastasis has occurred in the patient. Currently, the detection of metastasis is performed by pathologists, reviewing large extensions of biological tissues. This process is labor intensive and error prone.
The enterprise method leverages a convolutional neural network (CNN) architecture and obtains next generation results in the data set Camelyon16 in the difficult task of tumor detection at lesion level.
This high-level image recognition was first developed for Google's carless driver program to help vehicles to detect road obstructions.
If you would like to read the full Google article and download it, click here.
