HPC Industry Poised to Support BRAIN Initiative

BRAIN Initiative

According to researchers at MIT, 40 million Americans suffer from anxiety disorders, such as posttraumatic stress disorder, social phobias and obsessive-compulsive disorder. Experts say that treating these kinds of anxiety disorders can benefit from a better understanding of the brain circuits. While neuroscientists have taken major strides in understanding the human brain, the next step requires a multi-disciplinary approach where neuroscientists work side by side with researchers and computer scientists.

President Obama’s recently announced Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative, a National Institutes of Health (NIH) program aimed at improving our understanding of the human brain. The BRAIN Initiative is still in its infancy and it is difficult to define which organizations and industries will end up having a major impact on the project. However, one thing is clear – the unique capabilities of high-performance computing systems are a natural fit for the BRAIN Initiative’s research efforts.

One of the techniques utilized in brain research is called functional magnetic resonance imaging (fMRI), which is equivalent to dissecting the body into thin layers, allowing researchers to create 3-dimensional images of the brain that help diagnose diseases. Today, new techniques are being developed to visualize how neurons interact with each other. Recently, the Wall Street Journal has reported that scientists have been able to visualize the interaction of 1,000 neurons in a mouse for a period of a month.

What the BRAIN Initiative is trying to accomplish
Disorders of the brain have proven to be difficult to treat and medical professionals are hoping that the BRAIN Initiative will provide the insight needed to help care for patients more effectively. An article in the March issue of Science Magazine explained that brain research efforts have unlocked the secrets to how individual neurons operate on macroactivity within the brain. However, there are major gaps in identifying the real-time interactions between neurons during activities like thinking and memory tasks.

According to the NIH report detailing the BRAIN Initiative, the project is aimed specifically at this uncertain middle-ground in brain research. The overarching goal of the project is to develop more informative and detailed brain maps that show neural circuits and brain cells interacting in real-time.

This aim is an ambitious one that could enable the BRAIN Initiative to make a major mark on the scientific community. The NIH recognizes that reaching this end depends heavily on technological innovation. As a result, the initiative has established spurring technological development as a foundational part of the BRAIN project. The NIH is striving to not only develop new technologies that support BRAIN research, but also begin applying them to support better brain mapping.

Big data comes into play in just about any contemporary discussion pertaining to advanced research efforts. Analyzing and modeling operations within the brain can be mind-boggling, especially when trying to map how they take place in real time. Realistic modeling is essential in this area, and the task depends heavily on high-performance computing architectures to get the job done. HPC systems are emerging as major contributors to the big data movement because they are optimally suited to putting the large quantities of unstructured information into high-definition images and models that are essential for researchers.

Understanding the importance of such initiatives
Large-scale projects like BRAIN are becoming paramount to serve as catalysts for innovation. This is especially true in neuroscience, where meaningful advances require collaboration between engineers and scientists. Cray will be discussing these types of issues when it attends the upcoming Neuro Informatics 2013 Congress (INCF). We will be at the conference in Stokholm this week, and will be available to talk about the unique skills and tools needed to perform operations pertaining to visualizing interactions at both the molecular and behavioral levels.

BRAIN is not alone
Although the BRAIN initiative is just starting, the project can leverage the work of researchers at the McGovern Institute for Brain Research at MIT and other institutions that have been working on a similar effort. The Summer 2013 issue of Brain Scan magazine from the McGovern Institute explained that the institute has been using advanced technologies to promote better brain mapping. To a great extent, this project has only been successful because of effective interactions between the neuroscientists working at the facility and engineers involved in the effort.

Cray envisions the success at MIT and other institutions as a clear example of what the BRAIN Initiative needs to do in order to meet its goals. The BRAIN Initiative is looking at having to analyze and map approximately 100 billion neurons with 1 quadrillion connections. The scale of that analysis is incredible, and each synapse in the effort fires roughly 10 times per second. Advanced technologies are integral to achieving these results, and big data and high performance computing could end up proving essential to the project’s success.

Carlos P. Sosa, Chemistry and Life Sciences Segment Manager 

Carlos_Sosav1

Speak Your Mind

*

(won't be published)