In Simple Terms
Researchers have found a better way to watch living brain cells deep inside a mouse’s brain while leaving the layers above them intact. Their method can reveal cell activity nearly 2 millimeters down and collects far more light than earlier approaches using the same laser power. It opens a window onto deeper brain circuits, but it still cannot reach every part of the brain.
The Challenge Goes Beyond Depth
Light does not travel easily through a living brain to reach deep regions. It scatters as it passes through tissue, mixing the signals from the cells researchers want to observe with a background glow that is difficult to distinguish. In some experiments, scientists have therefore had to remove the tissue covering their target to get a clearer view. A new platform developed by a team at Rockefeller University offers a deeper look at neural activity in mice without removing the overlying cortex.
The Challenge Goes Beyond Depth
The work addresses two problems at once. The deeper light travels into brain tissue, the more it scatters, making it harder to tell where a fluorescent signal came from. But simply sending in more powerful light is not a safe fix: raising the laser’s average power can heat and damage tissue. Researchers need to get a useful signal without exceeding safe exposure limits.
Fluorescence microscopes use molecules that emit light when activated. In brain imaging, these signals help scientists track cell activity, including calcium changes inside neurons. But scattered light can create a background glow that obscures the signal of interest. The farther the cells are from the brain’s surface, the worse the problem becomes—and the harder it is to read the activity of individual cells.
Three-photon imaging offers a potential advantage: it produces less background fluorescence than two-photon imaging, making it, in principle, suitable for looking deeper. To make use of that advantage, however, the method needs laser pulses powerful enough to produce a measurable signal. This creates a trade-off between seeing deeper and protecting the tissue—a problem that could not be solved simply by increasing the energy or changing one component in isolation.
Reworking the Imaging System
Rather than treating the imaging device as a collection of fixed settings, the team led by Alipasha Vaziri, head of the Laboratory of Neurotechnology and Biophysics at Rockefeller University, reconsidered how the system’s components work together. The researchers named their platform TrUDI, short for ultra-deep three-photon imaging. Their results were published in Nature Methods in a study on recording groups of neurons at cellular resolution and at a depth of about 2 millimeters.
The team used computer models to guide changes to the laser, the optics, and the way the signal was collected. The goal was not just to reach the greatest possible depth. The researchers also needed to balance the properties of the excitation pulses, how light traveled through the system and was gathered, and the imaging speed and field of view. The design also accounted for safe laser exposure and the need to measure calcium changes accurately enough to distinguish individual cells.
The central idea is that no single part of the device determines its overall performance. The choice of laser has to match the optical system, while the scanning and recording methods must suit the timescale of neural signals. If these elements are not coordinated, useful photons can be wasted or unwanted illumination can increase—even when each component works well on its own. The team therefore optimized the system as a whole instead of relying on a familiar adjustment to one setting.
A Stronger Signal Without More Laser Power
The study’s most striking quantitative result was that TrUDI detected about 250 times as many fluorescence photons as previous methods when using the same laser power. That comparison matters because detecting more photons provides more information about the signal. The improvement did not come from simply raising the laser’s average power, which could heat the tissue or cause damage.
This does not mean that every signal in every region became stronger by the same amount, or that all parts of the brain can now be seen. Rather, the coordinated changes increased the amount of fluorescent light the system could collect under the tested conditions, making it possible to distinguish the activity of individual cells at a depth that had previously been difficult to reach. The quality of any useful reading still depends on where the imaging is done, the tissue’s optical properties, and safety limits.
The researchers also added a way to excite selected cells without continuously illuminating the entire field. With this approach, the laser is turned on only as the scan passes over cells that have been identified in advance, rather than illuminating the entire field continuously. Directing the excitation toward the intended cells helps make light use more selective, as part of a broader set of improvements designed to strengthen the signal while taking into account how much light reaches the tissue.
From a Previous Limit of 1.2 Millimeters to Nearly 2 Millimeters
Before this work, functional recordings using three-photon imaging reached depths of about 1.2 millimeters. That was short of the theoretical depth the technique could reach, which is more than 2 millimeters. TrUDI’s results show that coordinating the laser, optics, and recording system can bring imaging closer to that greater depth, rather than treating earlier performance as a limit that cannot be surpassed.
The team tested the platform in mice and recorded activity from groups of neurons in the cerebral cortex and hippocampus at a depth of about 2 millimeters. The field of view measured about 230 × 230 micrometers, and the recording rate reached 8.6 hertz. These figures help define what the achievement involved: the method did not merely produce a still image of deep anatomy. It tracked changing neural activity over a defined area and at a stated recording rate.
These details matter when assessing what the device can do. A field of view of 230 × 230 micrometers covers a relatively small patch of tissue, and a rate of 8.6 hertz sets how often data are collected. The technique should not be imagined as showing the whole brain at once or recording every neural event without limits. Instead, it allows researchers to examine groups of cells in deep locations, subject to spatial and temporal constraints that need to be considered when designing experiments.
Why It Matters to Preserve the Cortex
When studying a deep region requires removing the tissue above it, the target is left in a different anatomical context, which can make comparisons with an undisturbed brain more complicated. Imaging through the overlying layers gives researchers a way to observe some deep circuits while leaving the cortex intact. This does not eliminate every effect of an experimental procedure, but it expands the possibilities for studying activity without removing the covering tissue when light can reach the target.
This capability could help researchers understand how groups of neurons coordinate their activity across deep brain regions. Instead of limiting observations to the brain’s surface, scientists can follow cells in the cortex and hippocampus at greater depths and study changes in their activity over time within the available imaging area. The platform is a measurement tool, not an explanation of how neural circuits work, but it could provide data that help researchers ask more precise questions and design better experiments.
It is important to distinguish reaching a particular depth from being able to image every structure at that depth. A target’s shape and location, as well as the properties of the surrounding tissue, all affect whether it can be seen. Vaziri noted that parts of the striatum remain out of reach of this method, while several more distant deep nuclei require invasive approaches. So a depth of nearly 2 millimeters does not mean the device can image any brain target a researcher chooses.
What the Study Shows—and What Remains Uncertain
The work shows that deep imaging can be improved by better coordinating the components of a system, rather than merely driving the laser to higher power. In TrUDI, the properties of optical excitation were matched to the capabilities of the optical system and the method of collecting data, while signal quality and thermal, temporal, and spatial limits were taken into account. This approach increased the fluorescence detected and enabled cellular recordings at a depth of about 2 millimeters in living mouse brains.
But the results do not mean that this technique is now a general replacement for every method of brain imaging, or that it removes the need for invasive procedures in all experiments. Some deep targets remain beyond the demonstrated reach, parts of the striatum are inaccessible, and imaging is limited to a particular field of view and recording rate. The platform’s value lies in expanding the area researchers can reach with light while preserving the cortex above their targets—not in eliminating all technical limits.
The next step will be to use the tool to investigate specific neuroscience questions and determine how well it suits other regions and experimental setups. For now, the study demonstrates that carefully planned, integrated improvements can bring three-photon imaging closer to its theoretical limit, and shows that approach in mice. The current results alone cannot establish how well the method will work for deeper targets or different imaging arrangements.
The Bottom Line
A team at Rockefeller University’s Laboratory of Neurotechnology and Biophysics developed TrUDI to image neural activity deep in a mouse’s brain without removing the cortex above the target. By coordinating the laser, optics, and recording system, the researchers detected about 250 times as many fluorescence photons at the same laser power and recorded activity at cellular resolution nearly 2 millimeters deep, compared with about 1.2 millimeters for earlier recordings. Tests covered the cortex and hippocampus within a 230 × 230 micrometer field of view at 8.6 hertz. The advance broadens the tools available for studying the brain, but it cannot reach every deep region and does not remove the need for invasive methods in some locations.