Capturing Brain Activity at 100 Volumes Per Second: A Five-Minute Review With Associate Prof. Leilei Peng

What is the focus of this research? 

Our research developed dual-view Bessel two-photon projection microscopy (dv-B2PM), a fast three-dimensional fluorescence imaging technique designed to study neuronal activity in the brain at 100 volumes per second while maintaining sub-micron optical resolution. 

One of the major challenges in neuroscience is balancing imaging speed with the spatial resolution needed to observe activity at the level of individual synapses and dendrites. dv-B2PM bridges that gap by providing both high-speed volumetric imaging and synaptic-scale detail. 

Using the system, we captured rapid calcium activity across extended neuronal structures while preserving fine spatial resolution. By combining dv-B2PM with localized two-photon optical stimulation, we were able to observe fast and continuous spatiotemporal calcium activity extending from the apical dendrite to the soma. 

How did the collaboration between the University of Arizona and Stanford University develop? 

The collaboration began during my sabbatical leave to Stanford University as a visiting professor. During that time, we worked together on developing several advanced three-dimensional imaging approaches for neural tissue imaging. 

The University of Arizona team focused on engineering innovations that dramatically increased the speed of 3D imaging for weak neuronal signals occurring within extremely small spatial scales. We also developed image analysis methods capable of extracting high-resolution spatiotemporal neural activity data. 

The Stanford team contributed expertise in designing and conducting biological experiments that could rigorously demonstrate the capabilities of the imaging technology. Together, the two teams combined strengths in optical engineering, image analysis, neuroscience, and experimental biology to study neural signaling in brain tissue with unprecedented speed and detail. 

What is dv-B2PM, and what makes it different from existing imaging approaches? 

The central idea behind dv-B2PM is to simultaneously record two perpendicular projection views of the same imaging volume. The system uses two objectives positioned at 90 degrees relative to one another above the brain tissue. 

A Bessel two-photon beam scans the volume while a camera and a photomultiplier tube collect fluorescence projection image pairs from two orthogonal directions at the same time. 

This dual-view approach solves a common challenge in neuronal imaging. Structures that overlap in one projection can often be distinguished in the second view. Together, the paired projections compress the three-dimensional information while preserving the spatial context needed to trace signals along complex dendritic pathways. 

Most importantly, the technique avoids the image readout time required for traditional multi-layer 3D image stacks, enabling dramatically faster volumetric imaging without sacrificing critical information. 

What inspired the development of this new imaging system? 

The idea emerged during my sabbatical at Stanford while working in the laboratory of my collaborator, Dr. Ding. During my visit, I built a two-photon Bessel light-sheet microscope that was capable of imaging neuronal activity in brain tissue at 100 frames per second. 

During one discussion, Dr. Ding asked a simple but challenging question: Could we increase the speed from 100 frames per second to 100 volumes per second? 

That question motivated me to rethink the problem entirely. Instead of trying to push the hardware to acquire images faster through brute force, I focused on increasing the information throughout of the imaging process itself. 

The concept of dv-B2PM emerged from that approach. Rather than simply collecting more images, we developed a method that captures more useful three-dimensional information within each measurement, greatly improving the effectiveness of 3D imaging detection. 

What were the most exciting findings from the study? 

One of the most exciting outcomes was the ability to observe rapid and highly localized activity events across entire neurons that were previously inaccessible with conventional microscopy techniques. 

These events occur at spatial and temporal scales that have traditionally been difficult to capture. Our observations suggest that they may play an important role in the fundamental mechanisms that neurons use to process and compute information. 

What new scientific questions can researchers now investigate? 

The capabilities of dv-B2PM open the door to studying how dendrites perform complex nonlinear computations throughout their intricate three-dimensional structures. 

Neural signals often unfold on millisecond timescales while varying across micrometer and even sub-micrometer spatial dimensions. Existing imaging technologies have struggled to capture these processes comprehensively. By alleviating those limitations, dv-B2PM allows researchers to explore neuronal communication and computation with a level of detail that was previously out of reach. 

If dv-B2PM opens a new window into the brain, what are you most eager to look for through that window? 

We are particularly excited about using dv-B2PM to investigate circuit-level integration in neural tissue and to better understand the principles of neuronal integration. 

In other words, we want to study how neurons receive, integrate, and transmit information across complex neural networks. By capturing fast activity throughout entire neuronal structures, dv-B2PM provides a unique opportunity to observe these processes as they occur and gain deeper insight into how the brain computes and communicates information.