Reflection Matrix Microscopy Could Expedite Neuroscience Research

A South Korean research team has developed an optical microscope capable of maintaining spatial resolution and acquiring a microscopic “map” of neural network activity in brain tissue as it images through the width of an intact mouse skull. The device pairs hardware components with computational adaptive optics (AO); the latter was initially conceptualized to correct optical aberrations in ground-based astronomy.

Testing showed the ability of the microscope to capture two-photon fluorescence images of a dendritic spine of a neuron behind a mouse skull with a spatial resolution near the diffraction limit. The achievement, team members report, makes it possible to examine a mouse brain in its native states.

To image a mouse skull — useful in neuroscience research — strong speckle noise and image distortion detract from scientists’ ability to acquire high-quality optical images. The skull, as a result, must often be thinned or even removed to enable effective neural network/brain tissue study.

The researchers, working within the Institute for Basic Science (IBS) Center for Molecular Spectroscopy and Dynamics, have named the device a reflection matrix microscope. Where traditionally practiced confocal microscopy measures a reflection signal at only the focal point of illumination, thereby discarding the remaining out-of-focus light, the reflection matrix microscope perceives and records all scattered photons at positions beyond the focal point.

Siemens star resolution target underneath a highly aberrating medium was used as a test sample to be imaged (a). A conventional optical coherence microscopy image before aberration correction (b). An aberration-corrected image obtained using the reflection matrix microscope. Courtesy of IBS (c).

Led by Choi Wonshik, the researchers then demonstrated the ability to correct the scattered photons using an AO algorithm called closed-loop accumulation of single scattering (CLASS). The team developed the algorithm in 2017; by exploiting scattered light, the algorithm enabled the scientists to extract ballistic light and correct even severe optical aberration defects.

Noninvasive microscopy techniques for in vivo living tissue imaging, in which light passes through a turbid material, generate both ballistic photons and multiply scattered photons. Ballistic photons travel directly through an object without experiencing deflection. Random deflections of light passing through a material form multiple scattered photons, which appear as speckle noise in a reconstructed image. As the distance through which light propagates increases, so too does the ratio between the two types of photons. This obscures image information.

In addition to noise generation, optical aberration of ballistic light causes contrast reduction and image blur during the image reconstruction process. Bone tissues, with particularly complex internal structures, are apt to generate significant multiple light scattering and optical aberration during in vivo imaging processes.

Existing noninvasive methods include optical coherence microscopy (OCM) and two-photon microscopy. Three-photon microscopy, used successfully for neuronal and deep-tissue imaging, is another option, though one that is limited by a low laser repetition rate; the technique relies on an excitation window in the infrared range, which can damage living tissue during in vivo imaging.

The research team responsible for developing and introducing the reflection matrix microscope demonstrated the ability to directly combine the device with a conventional two-photon microscope. Adaptive optics, based on hardware components within the reflection matrix microscope itself, eliminated the aberration the two-photon microscope’s excitation beam endures, counteracting the aberration of the mouse skull. Many existing AO microscopy systems require bright, point-like reflectors or fluorescent objects to function as “guide stars” (consider the origins of adaptive optics), whereas the new microscope performs without fluorescent labeling and without dependence on a target’s physical structures. In addition, conventional two-photon microscopy is unable to resolve delicate structures, like a neuron’s dendritic spine, without completely removing brain tissue from the skull. The number of distinct aberration modes users can correct is also more than 10× that of conventional AO systems.

The schematic of the reflection matrix microscope that was developed by researchers at the IBS Center for Molecular Spectroscopy and Dynamics. The system makes use of confocal scanning and a Mach-Zehnder interferometer, similar to optical coherence microscopy. However, instead of confocal detection, interferometric images of reflected waves from the sample are measured using a camera. In addition, a spatial light modulator (SLM) is introduced to physically correct sample-induced wavefront distortion. (BS: beamsplitter; GMx/y: galvo mirror; DG: diffraction grating; sDM: spectral dichroic mirror; OL: objective lens.) Courtesy of IBS.

Next steps will involve minimizing the microscope’s form factor and increasing its imaging speed. The researchers aim to develop a label-free reflective matrix microscope with high imaging depths that is usable in clinical settings.

The research was published in Nature Communications (www.doi.org/10.1038/s41467-020-19550-x).