Measurement and correction of in vivo sample aberrations employing a nonlinear guide-star in two-photon excited fluorescence microscopy

Abstract

We demonstrate that sample induced aberrations can be measured in a nonlinear microscope. This uses the fact that two-photon excited fluorescence naturally produces a localized point source inside the sample: the nonlinear guide-star (NL-GS). The wavefront emitted from the NL-GS can then be recorded using a Shack-Hartmann sensor. Compensation of the recorded sample aberrations is performed by the deformable mirror in a single-step. This technique is applied to fixed and in vivo biological samples, showing, in some cases, more than one order of magnitude improvement in the total collected signal intensity.

Keywords: (170.3880) Medical and biological imaging; (180.4315) Nonlinear microscopy; (190.4180) Multiphoton processes; (220.1000) Aberration compensation; (220.1080) Active or adaptive optics.

Fig. 1

Schematic experimental setup used for aberration measurements and wavefront correction. Ti:S is the Ti:sapphire laser, GM are the galvanometric mirrors, L#, lenses; DM is the deformable mirror, M#, mirror, HM is a dichroic filter, TL is the microscope tube lens, F is BG39 filter, NL-GS is the nonlinear guide-star, WFS is the wavefront sensor, and PMT is the photo multiplier tube. The microscope output port is manually selected either for the PMT or for the WF sensor. For optimum usage of the DM, it was placed as close to normal incidence as the optics mounts allow it.

Fig. 2

(a, b, c) three different WFs of the excitation beam showing an RMS variation of up to 70% and (d, e, f) their corresponding generated WFs measured using the SH WFS. The measured NL-GS WF is similar in all three cases, with a maximum WF error variation of 7% (from 0.252µm to 0.271µm RMS). RMS: root mean square; PV: peak to valley. The planes (tilts) and spherical (focus) components of the WFs have been removed.

Fig. 3

Recorded wavefront maps of three different WFs generated from i) NL-GS produced inside the red-paint test-sample, ii) from 1 µm radii fluorescent beads, and iii) 0.28 µm. radii fluorescent beads. The RMS values are 0.042, 0.040, and 0.037 µm, respectively. The planes (tilts) and spherical (focus) components of the WFs have been removed.

Fig. 4

Recorded off-axis NL-GS at 8 equidistant locations from the on-axis position (average RMS WFE is ~0.120 µm). The 8 recorded positions were ~17.9 µm apart from the center (RMS WFE is ~0.114 µm). The FOV for this experiment was ~44.8µm. The planes (tilts) and spherical (focus) components of the WFs have been removed.

Fig. 5

Resulting system calibration correction applied to the excitation beam. Left panel excitation beam and right panel corrected beam using a closed loop configuration. The initial WFE was 1.31 µm. After the system was calibrated for coupling aberrations, the residual wavefront error was 0.007 µm. The planes (tilts) and spherical (focus) components of the WFs have been removed.

Fig. 6

Single frame images taken from an in vivo C. elegans sample. The imaged depths are 25 µm, 35 µm and 45 µm for the first, second and third rows respectively. The improvement factors with respect to the uncorrected case are 1.75 for (b) and 3.61 for (c); 1.90 (e) and 2.35 (f); 1.62 (h) and 2.02 (i). The gained improvement factor with respect to the coupling aberrations are 2.06 (c), 1.24 (f), and 1.24 (i). The plot profiles on the last column correspond to green line in each image. The red spot corresponds to the position where the NL-GS was measured. The WFS integration time was set to 800 ms for all the cases.

Fig. 7

Single frame images taken from an in vivo C. elegans sample. The imaged depths are 115 µm, 125 µm and 135 µm for the first, second and third rows respectively. The improvement factors with respect to the uncorrected case are 1.89 for (b) and 5.32 for (c); 1.58 (e) and 4.78 (f); 1.29 (h) and 9.1 (i). The gained improvement factor with respect to the coupling aberrations are 2.82 (c), 3.03 (f), and 7.08 (i). The plot profiles on the last column correspond to the green line in each image. The red spot corresponds to the position where the NL-GS was measured. The WFS integration time was set to 800 ms for all the cases.

Fig. 8

Single frame images taken from an in vivo C. elegans sample. The imaged depths are 105 µm, 115 µm, and 127 µm for the first, second, and third rows, respectively. The improvement factors with respect to the uncorrected case are 1.70 for (b) and 17.04 for (c); 2.15 (e) and 11.24 (f); 1.94 (h) and 22.59 (i). The gained improvement factor with respect to the coupling aberrations are 10 (c), 5.23 (f), and 11.66 (i). The plot profiles on the last column correspond to the green line in each image. The red spot corresponds to the position where the NL-GS was measured. The WFS integration time was set to 800 ms for all the cases.

Fig. 9

Single frame images taken from mouse brain slices expressing GFP in sparsely distributed neurons. The imaged depths are 10 µm and 40 µm for the first and second rows, respectively. The improvement factors with respect to the uncorrected case are 2.74 for (b) and 5.69 for (c); 1.98 (e) and 3.91 (f). The gained improvement factor with respect to the coupling aberrations are 2.08 (c) and 1.98 (i). The plot profiles on the last column correspond to the green line in each image. The red spot corresponds to the position where the NL-GS was measured. The WFS integration time was set to 1000 ms for all the cases.