The human brain is a puzzle hidden in an enigma wrapped in a mystery. We can describe the basic parts that make up a brain but there is a vast distance between describing something and knowing how it works.
If we want to image the brain as it goes about its business in a living, breathing creature, we have limited options. One is the electroencephalograph. Another option is positron emission tomography. And a third choice is magnetic resonance imaging.
Each of these tools is useful for its own purposes, but none of them can see the activity of individual neurons inside the living brain with full clarity. This is a vexing problem and one for which the National Science Foundation (NSF) is eager to find a solution. The NSF has found a willing partner at Cornell University in this quest to create technologies that will allow researchers to image the brain and the nervous system.
Cornell University has been awarded $9 million by the NSF to establish a neurotechnology hub whose mission is to create and develop new technologies for imaging neural activity in the brain and to share what they discover with the broader neuroscience world. The Cornell Neurotechnology Hub will be known as NeuroNex and the five co-principal investigators are Chris Xu (Applied and Engineering Physics), Joseph Fetcho (Neurobiology and Behavior), Mert Sabuncu (Electrical and Computer Engineering/Meinig School of Biomedical Engineering), Chris Schaffer (Meinig School of Biomedical Engineering), and Nilay Yapici (Neurobiology and Behavior).
Each brings a unique bit of expertise to the endeavor, and together they hope to take three important steps.
The NeuroNex team will start by adapting existing three-photon microscopes so they can be used on a variety of species. Three-photon microscopy has been employed effectively with mice to image neural activity in deep areas of the brain without causing damage to brain cells. Chris Xu and his collaborators will enable the same tool to be used with fruit flies, zebrafish and other species just as effectively.
Xu also hopes to design and build a new type of microscope that will combine with a novel form of illumination to create an imaging system that adapts to the sample being studied, with the goal of speeding up how quickly we follow neural activity. These imaging tools will be developed and utilized in a new space, the Laboratory for Innovative Neurotechnology at Cornell.
In addition to their reliance on foundational work started at Cornell by Professor Watt Webb years ago, these imaging tools share another characteristic: they will create enormous amounts of digital imaging data. And data, without the tools to make sense of it, is useless.
Of the five investigators, one is on the faculty of ECE. Mert Sabuncu officially started at Cornell Engineering in 2017 with a joint appointments in ECE and BME. As he was visiting campus prior to joining the faculty, Sabuncu came to see how extensive the neuroscience community was across the university.
“In conversations during my early visits at Cornell, I met a broad community of scientists studying the neural systems of diverse organisms that included fruit flies, zebrafish, mice and humans,” says Sabuncu. “I became excited about the large amounts of interesting and novel data these researchers are creating, as my research interests are in developing methods to handle, analyze and learn from such data.”
The challenges of dealing with massive amounts of data is exactly what Sabuncu has been working on in one form or another since his undergraduate years at the Middle East Technical University in Ankara, Turkey.
“Since I was a teenager, I knew that I wanted to be a scientist,” says Sabuncu. “I am a very curious person and an obsessive learner. As an undergraduate I had the chance to do research involving computer vision for radar data. This exposed me to the areas of image processing and computer vision and I was hooked.”
During his time in a postdoctoral research position at MIT and in a faculty position at Harvard’s A.A. Martinos Center for Biomedical Imaging, Sabuncu let his learning obsession run free and he took in everything he could find about a wide range of clinical research areas and genetics.
“I went to classes, I attended talks, I asked a million questions,” says Sabuncu. “The more I learned the more passionate I became about biomedical data and the things I could do with it.”
As a result of Sabuncu’s conversations during his early visits at Cornell, he was brought in to help create the NSF proposal for NeuroNex. “At Cornell, I’ll be developing novel algorithmic technologies for making sense of and exploiting large-scale biomedical data,” says Sabuncu. “NeuroNex is exciting to me because it will let me look at data from different animal models and at different scales. I am eager to inject our understanding of the imaging device and underlying biology into the data analysis tools we create.”
The imaging tools to be created at NeuroNex will be able to collect sophisticated data about the activity of many neurons over time along with information about the animal’s behavior at that time. Due to the complexity of the brain and the sheer number of neurons interacting in any given moment, using existing simple statistical analysis methods is likely to shed limited new light on the workings of the brain and nervous system.
“When you are looking at larger scales, you’re looking at thousands, potentially tens of thousands, of neurons activating simultaneously in complex patterns,” says Sabuncu. “You cannot get away with simple statistical analysis. You need to think about large-scale patterns that relate to the complex behavior these animals display.”
Sabuncu is thrilled to be part of this effort at Cornell. “I have always been drawn to the theoretical,” says Sabuncu. “And this is still true today. Yet, for me, mathematical abstractions are interesting in the way they relate to real world problems. I enjoy using my theoretical insights to devise practical solutions. Cornell gives me a great opportunity to do this with NeuroNex and within ECE, which is a world-class department that is a pioneer in marrying theoretical insights with real-world problems and has a unique focus on the interface between hardware and software.”
