Analyzation, Improvement, and “Ah-Ha’s!” in Neuroscience: the Neurophotonics Center’s 7th Annual Symposium Recap
Yesterday, the Neurophotonics Center’s 7th annual symposium took place on the Center for Computing and Data Sciences’ 17th floor, organized by faculty members Mike Economo and Brian DePasquale. With hundreds of attendees and 11 presenters, the topics were broken down into four sessions under the umbrella of “Machine Learning and Photonics in Neuroscience.”
“Our community got to hear about a lot of cutting-edge research that spanned the many ways in which researchers in neuroscience and imaging are interfacing with machine learning to accelerate our understanding of brain function,” said DePasquale. “The event was a huge success!”
The first session, entitled “Improving measurements,” featured presentations from BU Assistant Professor Lei Tian, Harvard University fellow Dushan Wadduwage, and Harvard professor Adam Cohen. In each presentation, the presenters discussed subject matter ranging from the challenges faced during mesoscale brain-wide imaging at high resolution, to encoding light as a means to improve large-volume imaging of small brains, such as in mice.
Session two, “Improving analyses,” featured two Zoom presentations by Kanaka Rajan of Harvard University and Carsen Stringer of the HHMI-Janelia Research campus. With Gal Mishne, of UCSD, finishing the second session with an in-person presentation on the dynamics of functional connectivity, the focus on neuroscience shifted from adeptly capturing data toward understanding what the data can tell researchers within the field.
During lunch, attendees were invited to peruse the many posters presented by student researchers posted along the 17th floor windows, showcasing a wide-range of dedicated study and results.
Session three, “Making sense of behavior,” kicked off following lunch, covered by presenters Bob Datta of the Harvard Medical School and Annegret Falkner of the Princeton Neuroscience Institute. Here, the presenters focused on the in vivo study of animals to understand their natural behaviors, as well as the neural dynamics of social dominance and defeat.
“Of course I found all the presentations intriguing, but I have a special spot in my heart for the neural basis of naturalistic and complex behaviors because such behaviors are so integral to our lived experience as animals,” shared DePasquale. “The impressive advances that these presenters, such as Bob Datta and Annegret Falkner, discussed, which have been fueled by exciting new developments in machine learning, feel like such a rapid and giant step beyond what our understanding was in the past, that it’s hard not to be dazzled by their work. I believe this rapid progress is an excellent example of what cleverly combined approaches from machine learning and imaging can bring to the study of neuroscience, and provides an hopeful picture for future rapid progress in other domains.”
In the final session, “Making sense of cells and circuits,” BU professor Brian Clearly, alongside Tony Zador of the Cold Spring Harbor Laboratory and Jeff Lichtman of Harvard University, presented on subjects concerning algorithmic tools and technologies, brain wiring, and connectomics. In all three presentations, the subjects well-matched the session topic of “making sense,” discussing how to study and understand the inner workings of the brain.
At the conclusion of the four sessions, a reception and subsequent poster session were held, where drinks were had, connections made, and ideas shared among colleagues from all around the education and industry fields.
“The biggest takeaway I had is that we still have a lot of work to do!” said DePasquale. “Our final session was especially filled with examples of the tidal waves of data our field is about to become overwhelmed by, and it is only through new and well-designed machine learning approaches do we have a hope to make sense of these data. A related point is the continuing importance of collaboration, which the symposium itself was undoubtedly an act of: investigators working in different domains — imaging, behavior, machine learning, etc. — need to continue to come together and discuss challenges and progress, because it’s 100-percent clear to me that substantive progress in understanding the brain will only come from the combined expertise of many, many individuals with complementary specialties.”
“I lost track of the number of ‘ah-ha!’ moments I had during the symposium when a speaker would discuss a machine learning approach they were applying to say, imaging, that could have profound impact on the study of behavior, and vice versa. I hope the attendees had those ah-ha moments as well!”