Thanks to a machine learning algorithm based on graphics processors, researchers in India are better able to understand and predict the connections between different regions of the brain.
This algorithm is named The real life (Structured, Accelerated, Linear Evaluation of the Filling) and was created thanks to research conducted by scientists from the Indian Institute of Science. Details about this topic are shown in the pages computational natural sciences.
It is able to analyze huge amounts of data generated from human brain scans obtained using diffuse MRI. Using ReAL-LiFE, researchers in India were able to analyze this data more than 150 times faster than current methods.
The brain contains “channels” called axons
Our brain is a very active organ, although at first glance it may seem otherwise. There, millions of neurons are activated every second, generating electrical impulses that travel in neural networks from one place in the brain to another through the channels that connect them. Varsha Srinivasan, as one of the study’s authors, explains that understanding these connections is critical to discovering the relationship between brain function and certain behaviors.
These tunnels are called hubs, and research dedicated to them has so far been difficult. Using diffusion MRI, the scientists were able to track the movement of the axon in order to create a comprehensive map of the brain’s fiber network. Of course, algorithms that tune the expected and detected signals helped. Although it has already been used before life (Linear Vise Evaluation), it relied on traditional processors, so the computation took longer.
A modified version of the algorithm called The real lifeIt depends on the graphics processors. It turns out not only to perform calculations faster, but also to predict how the examined person will behave or how he will complete a certain task. This means that by reading the connections predicted by the algorithm, the scientists were able to explain the differences in behavioral and cognitive test scores between the 200 participants in the experiment. In practice, this solution can help identify early signs of aging or the first symptoms of Alzheimer’s disease.
“Infuriatingly humble musicaholic. Problem solver. Reader. Hardcore writer. Alcohol evangelist.”