Nvidia’s breakthrough is all set to improve the way artificial intelligence research team has been trying to analyze brain tumors. Even though AI is a big-time innovation that holds big promises for all the fields including the medical profession; researchers have been facing certain issues when it comes to study tumors in the brain. The inherent problem is that the AI technique has been coming up with abnormal brain images that are quite uncommon.
Researchers from the Mayo Clinic, Nvidia and the MGH & BWH Center for Clinical Data Science have come together and are gearing up to present a paper that will highlight the use of generative adversarial networks (GANs) for creating synthetic brain MRI images. GANs comprise two artificial intelligence systems that are positioned against each other. One of the systems creates synthetic results within a particular category whereas the other system identifies fake results. Both these systems improve by working in competition with each other.
This major breakthrough in artificial intelligence research help the GANs to expand the data on which the researchers and doctors have to work, especially that is related to rare brain diseases. According to a senior research scientist at Nvidia, Hoo Chang Shin, diversity is highly important for the success when it comes to training the neural networks. However, the research becomes more critical because the medical imaging data is more often imbalanced. Shin further added that the number of normal cases is much more than abnormal cases. Even then, it is the abnormal cases that the medical science is more anxious to know about for accurate detection and diagnosis of a disease, says ZDNet.
The group of researchers led by Shin is to present this research on brain imaging at Spain’s MICCAI Conference that is going to explore the intersection of medical imaging and computer science. According to Shin and his colleagues, GANs will widen the potential data sets and provide a solution for the privacy challenges when it comes to using a patient data.
Incidentally, it is not just a medical profession, but even the military field that is using artificial intelligence research for betterment. According to Analytics India Magazine, the research arm of US military, Defense Advanced Research Projects Agency (DARPA) has decided to invest $2 billion for research and development in artificial intelligence for the next five years. The tweet on the official handle of DARPA even says that its “AI Next” campaign is aimed at accelerating AI’s third wave with the aim to increase trustworthiness between machines and humans.[The views and opinions expressed in this article are those of the authors and do not necessarily reflect the views and/or the official policy of the website. ]