Geometric transformation adaptive optics (GTAO) for volumetric deep brain imaging through gradient-index lenses

Abstract

The advance of genetic function indicators has enabled the observation of neuronal activities at single-cell resolutions. A major challenge for the applications on mammalian brains is the limited optical access depth. Currently, the method of choice to access deep brain structures is to insert miniature optical components. Among these validated miniature optics, the gradient-index (GRIN) lens has been widely employed for its compactness and simplicity. However, due to strong fourth-order astigmatism, GRIN lenses suffer from a small imaging field of view, which severely limits the measurement throughput and success rate. To overcome these challenges, we developed geometric transformation adaptive optics (GTAO), which enables adaptable achromatic large-volume correction through GRIN lenses. We demonstrate its major advances through in vivo structural and functional imaging of mouse brains. The results suggest that GTAO can serve as a versatile solution to enable large-volume recording of deep brain structures and activities through GRIN lenses.

Fig. 1. System design of GTAO and point spread function quantification.

a The implementation of GTAO via a 90-degree spatial profile rotation and the associated optical wavefronts. b The optical design of the GTAO-based two-photon microscope. A1 and A2, magnetic mount which allowed the insertion of alignment optics; Obj, objective lens; L1-10, optical lenses; DM, dichroic mirror; PMT, photomultiplier tube; F, emission filter. c The focal profiles observed under GRIN lens 2 by a water immersion objective lens and a camera. The experiment was repeated independently 5 times with similar results. d, e Two-photon fluorescence imaging of 1μm beads without and with GTAO, respectively. f–i The corresponding transverse and axial profiles of the bead images. The beads imaging was repeated independently 6 times with similar results. j, k The axial cross-sectional plot without and with GTAO, respectively. The beads imaging was repeated independently 3 times with similar results. l–o The comparison of peak intensity and spatial profiles between the images acquired with and without GTAO.

Fig. 2. The adaptability of GTAO to working distance, GRIN lens, and laser wavelength.

a The optical configuration for imaging at different working distances. b, c Two-photon images of 1 µm beads without and with GTAO at different working distances. The zoomed-in views of the beads in the dashed box were shown in the upper right corner. The beads imaging was repeated independently 4 times with similar results. d, e The ratio of peak intensity and axial FWHM in the outer regions of the FOV, respectively. f The optical configuration for using GRINTECH lens to correct Gofoton lens. g, h Two-photon images of 1 µm beads without and with GTAO, respectively. The beads imaging was repeated independently 2 times with similar results. i The optical configuration for using the correction pattern established for 930 nm at 1040 nm. j, k Two-photon imaging of 1 µm beads at 1040 nm without and with GTAO, respectively. The beads imaging was repeated independently 2 times with similar results.

Fig. 3. GTAO-based structural imaging.

a, b Maximum intensity projection (MIP) of the image stack of Thy1-eGFP mouse brain slice along the axial direction acquired without and with GTAO, respectively. c, d The axial cross-sectional images of two regions of interest (ROI) without and with GTAO, respectively. e The axial plot of ROI 1 and 2. f The contrast ratio over the imaging FOV (n = 3017 ROIs from one brain slice). The imaging experiment was repeated independently 5 times with similar results. g, h MIP of in vivo structural images of Thy1-eGFP mouse brain without and with GTAO, respectively. i Cross-sectional plot for ROI 3 and 4. j Zoomed-in view of ROI 5 (dendrites). k Cross-sectional plot along the dashed line in ROI 5. l Image contrast ratio over the FOV (n = 111 somata from 2 mice). The in vivo structural imaging was repeated independently in 5 mice with similar results.

Fig. 4. 3D image registration.

a, b Focal plane through GRIN lens with GTAO and through an objective lens, respectively. c Two-photon image stack of Thy1-eGFP brain slice through GRIN lens. d Two-photon image through GRIN lens after image processing. e The same image plane recorded through the objective lens. The Thy1-eGFP registration imaging was repeated independently 3 times with similar results.

Fig. 5. GTAO-based functional imaging.

a The galvo control for near-simultaneous comparison of calcium imaging without and with GTAO. b, c Two-photon calcium imaging without and with GTAO, respectively. (d) Calcium transients from ROI 1 and 2. e Statistics of the calcium transients in d. Lines and error bars represent the mean ± standard error of the mean. The scattered dot plot represents individual data values. Two-sided Wilcoxon matched-pairs signed rank test, ***P = 1.0089e-246, n = 1500 frames for ROI 1 and 2. f The ratio of the image contrast, n = 141 somata from 4 mice. g Configuration for GTAO-based volumetric calcium recording. h Calcium images at 70, 90, 130, and 190 µm depth. i The corresponding calcium transients from these four image planes in response to external stimulation. The in vivo calcium imaging was repeated independently in 6 mice with similar results.