Scientists can be better able to comprehend and forecast connections between various brain areas thanks to a new GPU-based machine learning algorithm created by researchers at the Indian Institute of Science (IISc).
The algorithm, known as Regularized, Accelerated, Linear Fascicle Evaluation or ReAl-LiFE, is capable of efficiently analyzing the massive volumes of data produced by diffusion magnetic resonance imaging (dMRI) scans of the human brain.
The team’s use of ReAL-LiFE allowed them to analyze dMRI data more than 150 times quicker than they could have with current state-of-the-art techniques.
How does the brain connectivity model work?
Every second, the brain’s millions of neurons fire, creating electrical pulses that move via neural networks—also known as “axons”—from one part of the brain to another.
For the brain to function as a computer, these connections are necessary. However, traditional methods for studying brain connections often involve using invasive animal models.
However, dMRI scans offer a non-invasive way to examine human brain connections.
The information highways of the brain are the cables (axons) that link its various regions. Water molecules travel along with axon bundles along their length in a directed manner since they are formed like tubes.
The connectome, which is a detailed map of the network of fibers spanning the brain, can be made possible by dMRI, which enables researchers to follow this movement.
Unfortunately, identifying these connectomes is not simple. Only the net flow of water molecules at each location in the brain is shown by the scans’ data.
Consider the water molecules as automobiles. Without knowing anything about the roadways, the only information collected is the direction and speed of the cars at each point in time and place.
By monitoring these traffic patterns, the task is comparable to inferring the networks of roadways. Conventional approaches closely match the expected dMRI signal from the inferred connectome with the actual dMRI signal in order to correctly identify these networks.
To do this optimization, scientists earlier created an algorithm called LiFE (Linear Fascicle Evaluation), but one of its drawbacks was that it operated on conventional Central Processing Units (CPUs), which made the computation time-consuming.
ReAl-LiFE is a revolutionary model that was created by Indian researchers
Initially, researchers created an algorithm called LiFE (Linear Fascial Evaluation) to do this adjustment, but one of its disadvantages was that it depended on ordinary Central Processing Units (CPUs), which took time to compute.
Sridharan’s team improved their technique in the newest study to minimize the processing work required in a variety of ways, including removing redundant connections and significantly improving LiFE’s performance.
The technology was refined further by the researchers by engineering it to work on Graphics Processing Units (GPUs), which are specialized electrical chips used in high-end gaming PCs.
This allowed them to examine data 100-150 times faster than previous approaches. This updated algorithm, ReAl-LiFE, could also anticipate how a human test subject will act or do a certain job.
In other words, using the algorithm’s projected link strengths for each individual, the team was able to explain variances in behavioral and cognitive test scores among a sample of 200 individuals.
Such analysis can also have medicinal uses.” Large-scale data processing is becoming increasingly important for big-data neuroscience applications, particularly in understanding healthy brain function and brain disorders.
Conclusion
In conclusion, ReAl-LiFE could also anticipate how a human test subject will act or do a certain job.
In other words, using the algorithm’s projected link strengths for each individual, the team was able to explain variances in behavioral and cognitive test scores among a sample of 200 individuals.
Such analysis can also have medicinal uses.” Large-scale data processing is becoming increasingly important for big-data neuroscience applications, particularly in understanding healthy brain function and brain disorders.
Leave a Reply