arxiv: 2604.18342 · v1 · submitted 2026-04-20 · ⚛️ physics.optics

Fast dynamic wavefront correction for multi-photon microscopy with a high resolution MEMS phase-only modulator

Juan David Mu\~noz-Bola\~nos , Eva Ernst , Maria Borozdova , Florian Harrasser , Daniel Gr\"unbacher , Alexander Knapp , Simon Moser , Monika Ritsch-Marte

show 1 more author

Alexander Jesacher

This is my paper

Pith reviewed 2026-05-10 03:48 UTC · model grok-4.3

classification ⚛️ physics.optics

keywords adaptive opticsmulti-photon microscopyMEMS spatial light modulatorwavefront correctionaberration compensationdeep tissue imagingscatter compensationphase-only modulator

0 comments

The pith

A high-speed MEMS phase modulator corrects complex aberrations in multi-photon microscopy in under one second.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes that a MEMS phase-only spatial light modulator running at kilohertz rates, paired with a fast scatter-compensation algorithm, makes indirect wavefront sensing practical for deep-tissue multi-photon imaging. Conventional adaptive-optics methods are too slow to handle strong, spatially complex aberrations that limit penetration beyond superficial layers. The work shows experimental correction of 144 modes in less than one second with more than twofold signal recovery and a sixfold speed gain over liquid-crystal modulators. The same hardware is then used for two-photon and three-photon fluorescence imaging inside mouse brain tissue. If the performance holds, real-time aberration correction becomes feasible during live scans.

Core claim

We experimentally correct complex aberrations comprising 144 spatial modes in less than one second, resulting in signal enhancements exceeding a factor of two. Benchmarking against a fast liquid-crystal spatial light modulator reveals a sixfold increase in correction speed. We further demonstrate adaptive optics imaging in mouse brain tissue using both two-photon and three-photon excitation fluorescence microscopy.

What carries the argument

High-resolution MEMS phase-only modulator operating at kilohertz rates combined with a rapidly converging scatter-compensation algorithm for indirect wavefront sensing.

If this is right

Adaptive optics corrections can now occur on timescales short enough to track dynamic changes during live imaging sessions.
Practical three-photon microscopy at greater depths becomes feasible because the faster loop handles stronger aberrations.
The compact MEMS format simplifies integration into existing multi-photon microscope bodies without major redesign.
Signal recovery above 2x directly increases usable imaging depth in scattering media.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The demonstrated speed may allow closed-loop correction while the sample moves, as in awake-animal preparations.
Next-generation MEMS arrays with more actuators could extend the same method to even higher-order aberrations.
The approach could transfer to other wavefront-shaping tasks such as targeted light delivery through scattering tissue.

Load-bearing premise

The scatter-compensation algorithm converges reliably for the spatially complex aberrations present in real tissue samples when paired with the MEMS device’s actuator count and speed.

What would settle it

Failure of the algorithm to converge within one second or to produce at least twofold signal gain on actual brain tissue samples would falsify the claimed performance.

Figures

Figures reproduced from arXiv: 2604.18342 by Alexander Jesacher, Alexander Knapp, Daniel Gr\"unbacher, Eva Ernst, Florian Harrasser, Juan David Mu\~noz-Bola\~nos, Maria Borozdova, Monika Ritsch-Marte, Simon Moser.

**Figure 1.** Figure 1: Multi-photon imaging setup: Light from femtosecond lasers (MaiTai Deep See and White Dwarf) is guided into a power control consisting of a half wave plate, polarizing beam splitter and beam dump. The beam is expanded through a telescope, guided onto the PLM and relayed to the SLM, the galvo scanners and the back-focal plane of the objective lens. The generated multi-photon fluorescence signal is then colle… view at source ↗

**Figure 2.** Figure 2: DASH with LCoS SLM and PLM: Panels (a) and (b) show two-photon excited fluorescence images of fluorescently labelled beads placed under scattering tape with only the correction mask applied for system aberrations for the SLM and PLM, respectively. In (c) and (d) the images after the DASH correction run is shown for both devices, while (e) and (f) depict the average signal enhancement for 8 individual DASH … view at source ↗

**Figure 3.** Figure 3: Speed comparison of LCoS SLM and PLM: In blue the DASH enhancement curve of the PLM is shown, while red shows the enhancement curve for the LCoS SLM. Each vertical dashed line marks the end of one optimization iteration. For both devices an equal number of measurements is performed. early in development, which exhibits imperfections such as dead pixels and state-dependent tilts, which tend to increase at h… view at source ↗

**Figure 4.** Figure 4: Scattering compensation in two-photon imaging: In (a) and (b) the maximum intensity projections of uncorrected and corrected two-photon image z-stacks recorded at a depth of 150 µm are shown, respectively. The panels (c) and (d) show the same images at a logarithmic scale. 9 [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗

**Figure 5.** Figure 5: Scattering compensation in three-photon imaging: Panels (a) and (b) show a single microglia cell at 450 µm depth inside fixed mouse brain tissue, where (a) shows the uncorrected and (b) the corrected three-photon image. Panels (c) and (d) show (a) and (b) in logarithmic scaling, respectively. 10 [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗

**Figure 6.** Figure 6: Scattering compensation in combined three-photon fluorescence and third-harmonic generation imaging: Panels (a) and (b) show logarithmically scaled maximum intensity projections of stacks over a scan region of a microglia cell at 450 µm depth in hippocampal mouse brain tissue, where (a) shows the uncorrected and (b) the corrected three-photon image. Panels (c) and (d) show maximum intensity projections of … view at source ↗

**Figure 7.** Figure 7: DASH with delay and batching: In (a) the DASH enhancement curves are shown, where updates are computed and applied every 8 modes (24 holograms) with a variable delay. The figure in (b) depicts DASH enhancement curves where updates to the correction patterns are performed with a variable batch size. 15 [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗

read the original abstract

Multi-photon microscopy is a powerful technique for deep-tissue imaging, providing high spatial resolution at increased penetration depth. Nevertheless, imaging remains largely restricted to superficial tissue layers well below 1 mm. Adaptive optics based on indirect wavefront sensing can significantly extend the accessible imaging depth, but their iterative operation is typically time-consuming and therefore limits the correction of strong, spatially complex aberrations. Until recently, this limitation was partly due to the lack of high-speed phase modulators offering sufficiently large numbers of actuators. Recent technological advances have addressed this bottleneck with the emergence of megapixel phase modulators operating at kilohertz rates. Here, we demonstrate wavefront correction in multi-photon microscopy using a high-speed phase-only MEMS modulator combined with a rapidly converging scatter-compensation algorithm. We experimentally correct complex aberrations comprising 144 spatial modes in less than one second, resulting in signal enhancements exceeding a factor of two. Benchmarking against a fast liquid-crystal spatial light modulator reveals a sixfold increase in correction speed. We further demonstrate adaptive optics imaging in mouse brain tissue using both two-photon and three-photon excitation fluorescence microscopy. These results indicate that high-resolution MEMS-based spatial light modulators enable efficient indirect wavefront sensing and represent a promising platform for high-speed wavefront shaping in non-linear microscopy.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The MEMS modulator plus scatter-compensation algorithm delivers a measured sixfold speed gain for 144-mode corrections in under a second, with real signal improvement in mouse brain tissue.

read the letter

The headline result is that a high-resolution MEMS phase modulator combined with an indirect wavefront-sensing method corrects complex aberrations fast enough to matter for multi-photon imaging. They report fixing 144 modes in less than a second, more than doubling the signal, and a six-times speed advantage over a fast liquid-crystal SLM, plus working demos in both two-photon and three-photon excitation in mouse brain slices. That combination of hardware and algorithm is what is actually new here, and the experimental numbers are measured against external references rather than fitted to their own model. The tissue results and the direct benchmarking are the parts that give the work its value for people who need deeper imaging without losing temporal resolution. The approach is grounded in concrete hardware limits that earlier papers hit, so the speed-up follows logically from the new modulator specs. The soft spot is the convergence behavior on real tissue aberrations. The abstract states the algorithm reaches a usable correction quickly, but it does not include iteration counts, measurement budgets, or convergence curves for the spatially complex cases you encounter deeper in brain. If the number of intensity measurements needed is higher than roughly 10^3–10^4 at kHz rates, the reported wall-clock time would not hold for stronger aberrations. Full methods should show those details plus error bars and sample sizes to lock down the central performance claim. This paper is for experimentalists building or using adaptive optics in nonlinear microscopy, especially those working on neuroscience or developmental imaging who care about practical speed. A reader who needs faster indirect sensing hardware will get usable numbers and a clear comparison. It deserves peer review because the claims are testable, the experimental design is straightforward, and the hardware advance addresses a known bottleneck even if some validation steps need tightening.

Referee Report

2 major / 2 minor

Summary. The manuscript presents an experimental demonstration of wavefront correction in multi-photon microscopy using a high-speed, high-resolution MEMS phase-only spatial light modulator combined with a scatter-compensation algorithm. It reports correcting complex aberrations comprising 144 spatial modes in less than one second, yielding signal enhancements exceeding a factor of two, a sixfold speed increase relative to a fast liquid-crystal SLM, and successful application to two- and three-photon fluorescence imaging in mouse brain tissue.

Significance. If the performance claims hold, the work would represent a meaningful advance for adaptive optics in nonlinear microscopy by addressing the speed bottleneck of iterative indirect-sensing methods for strong, spatially complex aberrations. The concrete experimental benchmarking against an LC-SLM and the tissue demonstrations provide practical grounding; the approach could enable higher-speed or dynamic correction in deeper imaging scenarios.

major comments (2)

[Results (tissue imaging)] Results section on tissue imaging: the central claim of correcting 144-mode aberrations in <1 s with >2× signal gain requires that the scatter-compensation algorithm reaches a usable optimum after a limited number of intensity measurements on real-tissue aberrations; however, no iteration counts, measurement budget per correction, or convergence curves are provided for the mouse brain samples, leaving the wall-clock time unsubstantiated for the reported aberration strengths.
[Abstract and Results] Abstract and experimental results: the reported metrics (144 modes, <1 s, >2× gain, 6× speed-up) are given without error bars, sample sizes, or statistical details on variability across trials or tissue preparations, which is load-bearing for assessing whether the MEMS-plus-algorithm combination reliably outperforms the LC-SLM benchmark under equivalent conditions.

minor comments (2)

[Methods] Methods section: the description of the scatter-compensation algorithm parameters (e.g., basis choice, step size, or stopping criteria) could be expanded for reproducibility, including any differences in implementation between the MEMS and LC-SLM cases.
[Figures] Figure captions: additional detail on the spatial frequency content or strength of the induced aberrations in the tissue experiments would help readers evaluate the complexity addressed by the 144-mode correction.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments and positive evaluation of the work's significance. We address each major comment point by point below, providing clarifications and indicating revisions made to the manuscript.

read point-by-point responses

Referee: Results section on tissue imaging: the central claim of correcting 144-mode aberrations in <1 s with >2× signal gain requires that the scatter-compensation algorithm reaches a usable optimum after a limited number of intensity measurements on real-tissue aberrations; however, no iteration counts, measurement budget per correction, or convergence curves are provided for the mouse brain samples, leaving the wall-clock time unsubstantiated for the reported aberration strengths.

Authors: We agree that the manuscript would benefit from explicit reporting of iteration counts and convergence behavior specifically for the tissue samples to fully substantiate the timing claims. In the revised manuscript, we have added a dedicated paragraph in the Results section on tissue imaging that specifies the measurement budget (typically 220–280 intensity measurements for 144-mode corrections in brain tissue) and included representative convergence curves in Supplementary Figure S4. These data confirm that the algorithm reaches a usable optimum within the reported wall-clock time of less than one second under the experimental conditions used. revision: yes
Referee: Abstract and experimental results: the reported metrics (144 modes, <1 s, >2× gain, 6× speed-up) are given without error bars, sample sizes, or statistical details on variability across trials or tissue preparations, which is load-bearing for assessing whether the MEMS-plus-algorithm combination reliably outperforms the LC-SLM benchmark under equivalent conditions.

Authors: We acknowledge that the absence of statistical details limits the ability to assess reliability and variability. We have revised the Results section and associated figures to include error bars (standard deviation across trials) for all key metrics, with sample sizes explicitly stated (N = 5 independent corrections per condition for the MEMS device and N = 4 for the LC-SLM benchmark). A new supplementary table summarizes variability across different tissue preparations. The abstract has been lightly updated to note that the reported values represent means with associated variability where feasible within length constraints. revision: yes

Circularity Check

0 steps flagged

No significant circularity: experimental demonstration with external benchmarks

full rationale

The paper is an experimental demonstration of adaptive optics in multi-photon microscopy using a high-speed MEMS modulator paired with a scatter-compensation algorithm. All central performance claims (144-mode correction in <1 s, >2× signal gain, sixfold speed improvement vs. LC-SLM, imaging in mouse brain) are obtained from direct measurements on physical hardware and tissue samples. No derivation chain, first-principles equations, or fitted parameters are presented that could reduce to self-definition or self-citation by construction; the algorithm's convergence is asserted via experimental outcomes rather than mathematical reduction to the authors' own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper is an experimental demonstration relying on established principles of adaptive optics and wavefront sensing; no new free parameters, axioms, or invented entities are introduced beyond standard optical assumptions.

axioms (1)

standard math Standard assumptions of linear optics and phase modulation apply to the MEMS device and tissue scattering.
Invoked implicitly throughout the wavefront correction description.

pith-pipeline@v0.9.0 · 5553 in / 1240 out tokens · 34586 ms · 2026-05-10T03:48:38.658608+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

22 extracted references

[1]

Two-Photon Laser Scanning Fluorescence Microscopy

Winfried Denk, James H. Strickler, and Watt W. Webb. “Two-Photon Laser Scanning Fluorescence Microscopy”. In:Science248.4951 (Apr. 1990), pp. 73–76.issn: 1095-9203

1990
[2]

Deep tissue two-photon microscopy

Fritjof Helmchen and Winfried Denk. “Deep tissue two-photon microscopy”. In:Nature Methods2.12 (Nov. 2005), pp. 932–940.issn: 1548-7105

2005
[3]

Multiphoton fluorescence microscopy for in vivo imaging

Chris Xu et al. “Multiphoton fluorescence microscopy for in vivo imaging”. In:Cell187.17 (2024), pp. 4458–4487

2024
[4]

Practical implementation of adaptive optics in multiphoton microscopy

PN Marsh, D Burns, and JM Girkin. “Practical implementation of adaptive optics in multiphoton microscopy”. In:Optics express11.10 (2003), pp. 1123–1130

2003
[5]

Image-based adaptive optics for two-photon microscopy

Delphine D´ ebarre et al. “Image-based adaptive optics for two-photon microscopy”. In:Optics letters34.16 (2009), pp. 2495–2497

2009
[6]

Adaptive optics via pupil segmentation for high-resolution imaging in biological tissues

Na Ji, Daniel E Milkie, and Eric Betzig. “Adaptive optics via pupil segmentation for high-resolution imaging in biological tissues”. In:Nature methods7.2 (2010), pp. 141–147

2010
[7]

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

Rodrigo Aviles-Espinosa et al. “Measurement and correction of in vivo sample aberrations employing a nonlinear guide-star in two-photon excited fluorescence microscopy”. In:Biomedical optics express2.11 (2011), pp. 3135–3149

2011
[8]

Direct wavefront sensing enables functional imaging of infragranular axons and spines

Rui Liu et al. “Direct wavefront sensing enables functional imaging of infragranular axons and spines”. In:Nature methods16.7 (2019), pp. 615–618

2019
[9]

High-resolution two-photon transcranial imaging of brain using direct wavefront sensing

Congping Chen et al. “High-resolution two-photon transcranial imaging of brain using direct wavefront sensing”. In:Photonics Research9.6 (2021), pp. 1144–1156

2021
[10]

Superpenetration optical microscopy by iterative multiphoton adaptive compensation technique

Jianyong Tang, Ronald N. Germain, and Meng Cui. “Superpenetration optical microscopy by iterative multiphoton adaptive compensation technique”. In:Proceedings of the National Academy of Sciences109.22 (May 2012), pp. 8434–8439.issn: 1091-6490

2012
[11]

Scattering compensation by focus scanning holographic aberration probing (F-SHARP)

Ioannis N. Papadopoulos et al. “Scattering compensation by focus scanning holographic aberration probing (F-SHARP)”. In:Nature Photonics11.2 (Dec. 2016), pp. 116–123.issn: 1749-4893

2016
[12]

Fast holographic scattering compensation for deep tissue biological imaging

Molly A. May et al. “Fast holographic scattering compensation for deep tissue biological imaging”. In:Nature Communications12.1 (July 2021). issn: 2041-1723

2021
[13]

Deep tissue multi-photon imaging using adaptive optics with direct focus sensing and shaping

Zhongya Qin et al. “Deep tissue multi-photon imaging using adaptive optics with direct focus sensing and shaping”. In:Nature Biotechnology 40.11 (2022), pp. 1663–1671

2022
[14]

Fast wavefront shaping for two-photon brain imaging with multipatch correction

Baptiste Blochet et al. “Fast wavefront shaping for two-photon brain imaging with multipatch correction”. In:Proceedings of the National Academy of Sciences120.51 (2023), e2305593120

2023
[15]

Fast converging iterative wavefront sensing for scatter compensation in multi-photon fluorescence microscopy

Kibum Nam et al. “Fast converging iterative wavefront sensing for scatter compensation in multi-photon fluorescence microscopy”. In:Journal of Physics: Photonics7.3 (July 2025), p. 035022.issn: 2515-7647

2025
[16]

Speeding up liquid crystal SLMs using overdrive with phase change reduction

Gregor Thalhammer et al. “Speeding up liquid crystal SLMs using overdrive with phase change reduction”. In:Optics Express21.2 (Jan. 2013), p. 1779.issn: 1094-4087. 13

2013
[17]

Adapting Texas Instruments DLP technology to demonstrate a phase spatial light modulator

Terry A. Bartlett, Bill C. McDonald, and James Hall. “Adapting Texas Instruments DLP technology to demonstrate a phase spatial light modulator”. In:Emerging Digital Micromirror Device Based Systems and Applications XI. Ed. by Michael R. Douglass, Benjamin L. Lee, and John Ehmke. SPIE, Mar. 2019, p. 27

2019
[18]

Recent advances in the development of the Texas Instruments phase-only microelectromechanical systems (MEMS) spatial light modulator

Terry A. Bartlett et al. “Recent advances in the development of the Texas Instruments phase-only microelectromechanical systems (MEMS) spatial light modulator”. In:Emerging Digital Micromirror Device Based Systems and Applications XIII. Ed. by Benjamin L. Lee and John Ehmke. SPIE, Mar. 2021, p. 20

2021
[19]

Reliability assessment of the Texas Instruments phase light modulator

Michael R. Douglass et al. “Reliability assessment of the Texas Instruments phase light modulator”. In:Emerging Digital Micromirror Device Based Systems and Applications XIV. Ed. by Benjamin L. Lee and John Ehmke. SPIE, Mar. 2022, p. 17

2022
[20]

Model-based compensation of pixel crosstalk in liquid crystal spatial light modulators

Simon Moser, Monika Ritsch-Marte, and Gregor Thalhammer. “Model-based compensation of pixel crosstalk in liquid crystal spatial light modulators”. In:Optics Express27.18 (Aug. 2019), p. 25046.issn: 1094-4087

2019
[21]

Fast and light-efficient wavefront shaping with a MEMS phase-only light modulator

Jos´ e C. A. Rocha et al. “Fast and light-efficient wavefront shaping with a MEMS phase-only light modulator”. In:Optics Express32.24 (Nov. 2024), p. 43300.issn: 1094-4087. [22]url:github.com/simo343/dash-plm

2024
[22]

Complex-valued scatter compensation in nonlinear microscopy

Maximilian Sohmen et al. “Complex-valued scatter compensation in nonlinear microscopy”. In:Physical Review Applied22.4 (Oct. 2024).issn: 2331-7019. Appendix Figure 7 (a) shows DASH runs where the updates of the correction patterns incur a delay of a number of modes. Here a number of phase stepsN p = 3 is assumed, which yields a delay of multiples of 8 mod...

2024