MIT Scientists Achieve Deeper Brain Imaging Using Photoacoustic Microscopy

Researchers at MIT have developed a new microscopy system that enables exceptionally deep imaging of brain tissues, capturing the molecular activity of individual cells using sound waves. This advancement helps address a major challenge in neuroscience: visualizing cellular processes not only in the brain's cortex but also in deeper regions like the hippocampus, which are otherwise difficult to access with traditional imaging methods.

Published in Light: Science, and Applications, the study demonstrates the system’s ability to detect NAD(P)H—a molecule closely linked to cell metabolism and neuronal electrical activity—at depths exceeding 1mm in dense specimens such as mouse brain slices and three-dimensional “mini-brains” grown from human stem cells. According to co-lead author W. David Lee, the imaging depth was constrained only by the sample size, not the system's capabilities.

The system combines several advanced techniques, allowing single-cell resolution at depths more than five times greater than traditional imaging technologies. It excites NAD(P)H in cells using an intense, ultra-short burst of light at three times the usual absorption wavelength, a process called “three-photon” excitation. This light penetrates tissue more efficiently due to its longer wavelength, thereby reducing scattering and providing clearer images at greater depths. While this excitation produces a weak fluorescent signal, the bulk of the absorbed energy generates small, localized thermal expansions in the cells. These expansions emit sound waves, which are captured by a sensitive ultrasound microphone. Software then turns the resulting sound data into high-resolution images, utilizing a process known as three-photon photoacoustic imaging.

Co-lead author Tatsuya Osaki described how the team merged three-photon, label-free, and photoacoustic techniques into a unified “Multiphoton-In and Acoustic-Out” approach. The system also supports third-harmonic generation imaging, which can visualize cellular structures, and may detect other molecules such as GCaMP, commonly used by neuroscientists as indicators of neural electrical activity.

This label-free, multiphoton, photoacoustic microscopy (LF-MP-PAM) has the potential for wide research and clinical applications. Because it does not rely on added chemicals or genetically modified markers, it is suitable for use in human procedures such as brain surgeries. NAD(P)H levels vary in conditions like Alzheimer’s disease and seizures, making this capability not only valuable for fundamental research but also for medical diagnostics.

The next challenge for the team is to adapt the technology for imaging in living animals, requiring new arrangements of the sound and light sources. The researchers are optimistic that imaging at even deeper levels—up to 2mm within live brains—will be possible, further expanding the utility of their technology for neuroscience and medicine.