Pith. sign in

REVIEW 3 major objections 5 minor 56 references

Compressive Fourier-Domain Intensity Coupling (C-FOCUS) enables near-millimeter deep imaging in the intact mouse brain in vivo

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read C-FOCUS claims that two-photon microscopy with Fourier-domain intensity masks computed by compressive sensing can image neurons and vasculature beyond 900 µm deep in the intact mouse brain, and through the intact skull, in vivo.

desk verdict Genuine first demonstrations in vivo with real anatomical evidence, but the linear model behind the method is undefended and the enhancement metric is circular; referee it and push for validation. read the letter →

arxiv 2505.21822 v1 pith:DPVCFERQ submitted 2025-05-27 physics.optics eess.IV

Renzhi He , Yucheng Li , Brianna Urbina , Jiandi Wan , Yi Xue This is my paper
classification physics.opticseess.IV PACS 42.25.Fx42.30.Wb87.64.M
keywords two-photonmicroscopyscatteringcorrectioncompressivesensingFourier-domainintensitymodulationdigitalmicromirrordevicedeepbrainimagingtranscranialinvivomouse
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper introduces C-FOCUS, a scattering-correction method for two-photon microscopy that uses a digital micromirror device to project intensity patterns on the Fourier plane, with the correction mask recovered from undersampled fluorescence measurements via compressive sensing. The authors claim this extends in vivo imaging depth beyond 900 µm in the intact mouse brain—resolving single axons in layer 6 and white matter and capillaries down to 1.5 µm—and enables transcranial imaging of dendrites through a 110 µm-thick intact adult skull. These results matter because conventional two-photon microscopy, even with adaptive optics, is typically limited to a few hundred micrometers, and the method avoids the iterative, slow phase-retrieval loops of most scattering-correction schemes.

What carries the argument

The load-bearing identity is the linear forward model $I = A\hat{x}$ in Eq. (1), where $I$ is the measured two-photon fluorescence under random binary mask patterns $A$ and $\hat{x}$ is the grayscale correction mask over 100 × 100 superpixels; the mask is recovered by solving a total-variation-regularized inverse problem with FISTA and then binarized to a DMD output-to-input power ratio near 29%. This model lets the system infer which Fourier-domain beams are in phase at the target and should be kept, versus blocked, using only 2,000–5,000 random measurements instead of 10,000. Content-aware subregions—formed by merging grid patches whose intensity peaks are too dim or too close—extend correction beyond the memory-effect range, and the correction masks are projected in synchrony with galvanometer scanning so the whole field of view is imaged in one seamless pass with constant surface power.

What would settle it

Image a single fluorescent bead through a fixed scattering layer, measure the fluorescence under the 3,000 random patterns, and compute a correction mask via Eq. (1). Then independently measure the fluorescence for a sequence of masks that linearly interpolate between the blank mask and the predicted mask at constant input power; if the peak fluorescence does not track the linear prediction—or if a mask built from the cross-terms alone produces comparable enhancement—the linear model is falsified and the reported depth gains could be due to trivial power redistribution rather than constructive interference.

Watch

Extended reading notes

Core claim

The central claim is that constructive interference among multiply scattered beams, selected by binary intensity masks placed on the Fourier plane, can be recovered from a small fraction of the measurements that full sampling would require, and that this is sufficient for in vivo deep-brain two-photon imaging. Using 1035 nm excitation, C-FOCUS produced 5- to 6-fold fluorescence enhancement in densely labeled YFP neurons at 800–900 µm depth, 12- to 24-fold enhancement in FITC-labeled vasculature at 700–910 µm depth, and 6- to 8-fold enhancement for apical dendrites imaged through the intact skull. The paper reports that the complete correction workflow, including data acquisition, transfer, computation, and mask projection, takes about 1.5 minutes for tens of subregions across a 126 × 126 µm² field of view, made possible because compressive sensing requires only 20–30% of the measurements of the prior 2P-FOCUS method.

Load-bearing premise

The load-bearing premise is that Eq. (1), $I = A\hat{x}$, correctly describes how much two-photon fluorescence each binary Fourier-plane intensity pattern produces, even though two-photon emission is quadratic in the excitation field and dominated by cross-terms between superpixels, so the linear relation is an idealization with no derivation in the paper.

Editorial extensions

If this is right

  • Correction time per field of view drops to about 90 seconds for 15 subregions, since compressive sensing cuts the raw measurement count to 20–30% of full sampling.
  • Imaging depth in the intact mouse brain extends past 900 µm at 1035 nm excitation, a range previously associated with three-photon microscopy rather than two-photon.
  • Transcranial two-photon imaging through an intact adult skull becomes feasible, eliminating the need for skull thinning or cranial windows in some experiments.
  • The method boosts fluorescence intensity by over 20-fold in vivo compared with uncorrected imaging, which the paper attributes to constructive multi-beam interference rather than simple NA reduction.
  • The technique is presented as compatible with longer excitation wavelengths, red-shifted fluorophores, three-photon microscopy, and spatially selective two-photon optogenetics.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the linear model $I = A\hat{x}$ holds quantitatively, the same compressive-sensing recovery could be combined with phase modulation (adaptive optics) to correct both intensity and phase, potentially improving resolution at depth rather than only brightness.
  • The content-aware subregion segmentation is a general strategy that could be reused in other multipatch wavefront-shaping schemes, since it avoids redundant masks and reduces the number of patches needed for a given field of view.
  • A testable extension is to apply C-FOCUS to tissues with different scattering statistics, such as tumor xenografts or spinal cord white matter, where the optimal measurement count and binarization threshold would need to be recharacterized.
  • The claim that the ultimate depth limit is set by the transport mean free path suggests that combining C-FOCUS with longer-wavelength excitation could push two-photon imaging to depths currently reached only by three-photon systems.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The manuscript presents C-FOCUS, a two-photon microscopy method that uses compressive sensing with random binary intensity patterns on a DMD at the Fourier plane to compute correction masks that enhance fluorescence from targets deep in scattering tissue. The authors demonstrate in vivo imaging of YFP-labeled neurons at 900 µm depth in the intact mouse brain through a cranial window, FITC-labeled blood vessels at 910 µm depth, and dendritic structures through an intact 110 µm-thick skull, with fluorescence enhancements up to 61-fold in ex vivo bead samples and 2.3- to 23.7-fold in vivo. The paper claims that C-FOCUS works by selecting in-phase multi-scattered beams to constructively interfere at the target, and that compressive sensing reduces the number of measurements to 20-30% of full sampling.

Significance. If the empirical results hold, C-FOCUS would extend two-photon microscopy well beyond the typical 500-800 µm depth limit and enable transcranial imaging without skull thinning, which is a substantial technical advance. The manuscript provides data and code availability, and the ex vivo PSF measurements (Fig. 2g-m) are direct evidence that the method can produce a focused spot through a scattering skull. However, the significance is tempered by two issues: the physical model underlying the correction-mask computation is not derived and appears inconsistent with the quadratic two-photon response, and the in vivo enhancement is evaluated on the same targets used for mask optimization. These issues must be resolved before the mechanistic claims and the quantitative improvement can be accepted.

major comments (3)
  1. [Section 2.1, Eq. (1)] The linear model I = Ax is introduced without derivation. For a single target, two-photon fluorescence under a binary mask s is proportional to |Σ_j s_j h_j|^4, which is quadratic in the intensity contributions and contains pairwise interference terms. Equation (1) models each measurement as a linear function of the mask weights, so it cannot determine the relative phases needed to select 'in-phase beams' as claimed in Section 2.1. A linear inversion over random intensity patterns can at best recover average per-superpixel intensity contributions. The manuscript provides no simulation, calibration, or argument demonstrating that the linear approximation holds under the experimental conditions (100×100 superpixels, 40% sparsity, 3000 measurements). This is load-bearing because every correction mask is computed from Eq. (1). The authors should either derive a regime in which the linear model is valid (or at least empirically validate it, e.g., by comparing measured fluorescence for new masks with the linear prediction), or revise the physical explanation to match what the linear model can actually recover.
  2. [Section 2.1 and Figs. 3-5] The fluorescence enhancement is measured on the same targets that were used to optimize the correction masks. In the content-aware subregion scheme, the uncorrected image is segmented and a local intensity peak is selected as the correction target for each subregion; the corrected image is then compared with the uncorrected image at the same subregions. The reported 2.3- to 61.3-fold gains thus represent the success of the optimization on its training objective, not an independent validation of scattering correction. At minimum, the authors should include a control in which masks are generated from one set of targets and the enhancement is measured on a different set of targets within the same or a neighboring FOV, or use a held-out subset of random patterns to verify predictive power.
  3. [Abstract and Section 3] The abstract states that C-FOCUS enhances fluorescence intensity 'by over 20-fold compared to uncorrected imaging,' but the representative in vivo results range from 2.3-fold (Fig. 3f) to 23.7-fold (Fig. 4b), with 5-fold in Fig. 1d and up to 6-fold in Fig. 3i. The unqualified claim in the abstract is not representative of the in vivo data except for the sparse-vasculature case at 700 µm. Please qualify the enhancement claim by stating the specific imaging condition, or present a median/range across all in vivo experiments.
minor comments (5)
  1. [Section 2.3] The text refers to 'Extended Data Fig. a-b' without numbers; it should be Extended Data Fig. 1a-b.
  2. [Discussion] There is a typo: 'CO-FOCUS' should be 'C-FOCUS' in the sentence 'CO-FOCUS corrects for the average scattering profile...'.
  3. [Methods] The typical TV regularizer weight is given as α = 0.008, but Extended Data Fig. 2f reports maximum enhancement at α = 0.006; please clarify which value is used for the in vivo experiments.
  4. [Section 2.2.1] The text says 'the results show that fluorescence intensity increases with the number of measurements, reaching a maximum 38.3-fold enhancement with 10,000 measurements' and later 'With C-FOCUS, the minimum improvement is 6.2-fold with only 10 measurements, and the maximum is 37.5-fold with 5,000 measurements.' The numbers are confusing; please clarify that the 38.3-fold is for 2P-FOCUS with 10,000 measurements while the 37.5-fold is for C-FOCUS with 5,000 measurements.
  5. [Figure 2 caption] The caption describes the overlap of correction masks from 2P-FOCUS (cyan) and C-FOCUS (green, magenta), but the colors in the actual figure should be checked to ensure they match the text.

Circularity Check

2 steps flagged · score 4.0 of 10

Enhancement and PSF metrics are measured on the very targets used to fit the correction masks, so part of the quantitative validation is the objective of the optimization; the linear model itself is imported from the authors' prior 2P-FOCUS work.

  1. fitted input called prediction [Section 2.2.1; Eq. (1) in Section 2.1; Fig. 2a]
    "The correction mask (x̂) is estimated by solving the inverse problem with a total variation (TV) regularizer: L(x̂; I) = ∥I − A x̂∥2² + α R_TV(x̂), (1) ... We acquired 10,000 measurements and generated correction masks from subsets of 10 to 5,000 measurements, which were then applied to image the same bead and evaluate fluorescence intensity (Fig. 2a)."

    The correction mask x̂ is solved from the PMT intensity vector I recorded at that same bead, and the reported enhancement is then measured on the same bead under the fitted mask. The 6.2- to 38.3-fold gains are therefore the success of fitting x̂ to I, not an independent test of the physical model: any mask that raises the target intensity will appear as 'enhancement.' The same structure recurs in vivo: subregion targets are local intensity peaks selected from the uncorrected image, masks are fit to those targets, and the improvement is quantified on those very targets (Fig. 3f-i, 4b-d, 5b-d).

  2. fitted input called prediction [Section 2.2.2; Fig. 2g]
    "To determine the optimal point, we imaged a single fluorescent bead (0.71 µm) through the same mouse skull (Fig. 2g) using correction masks generated with varying binarization thresholds."

    The bead whose image defines the point-spread function is the very bead used to generate the correction mask. The reported resolution (2.1 × 2.1 × 17.2 µm³) is therefore measured on the optimization target; it does not independently establish that the mask creates a comparable focus for other structures, because it only characterizes the effect of the mask on the object used for fitting. The paper later presents this as 'high-resolution imaging' and 'maintaining high resolution' for in vivo structures, but the only quantitative PSF is self-referential.

full rationale

The paper is an experimental demonstration rather than a first-principles derivation. The correction mask x̂ is obtained by fitting the linear model in Eq. (1) to PMT intensities I recorded at the target, and the principal quantitative metrics (fluorescence enhancement, PSF/resolution) are then measured on the same target used for that fit. These metrics are therefore not independent predictions of the model; they are the outcome of optimizing a mask for that target, which is a genuine but partial circularity (pattern 2). The physical rationale of 'in-phase beams' is inherited from the authors' own 2P-FOCUS paper [27], and the random-pattern/superpixel parameters are likewise set by that self-citation; no independent derivation of the linear intensity-coupling model is provided. However, the central claims—in vivo imaging at 720–910 µm depth, white-matter axons, and transcranial dendrites—rest on actual corrected images that show resolvable morphology (e.g., 1.1–2 µm axons, 1.5 µm capillaries) across multiple animals and depths. These observations are not mathematical consequences of the fit; they are empirical outcomes, and in some cases structures appear after correction where the uncorrected image at that depth showed none (Fig. 3e-i). Thus the depth and imaging claims retain independent content, and the circularity is partial rather than total.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The method depends on several tuned parameters and unverified physical assumptions. The main tuned parameters are the binarization threshold and the TV weight, both fitted to bead data or in vivo measurements. The most important assumption is the linear model in Eq. (1), which is not derived from the nonlinear two-photon response. No new physical entities are introduced.

free parameters (5)
  • DMD binarization threshold (output-to-input power ratio) = 29% (30% in most in vivo figures)
    Chosen by scanning thresholds from 7% to 43% on a bead through skull (Fig. 2g-m) to maximize intensity and resolution; all in vivo masks use this value.
  • TV regularizer weight alpha = 0.008 (best 0.006)
    Hyperparameter in Eq. (1); varied in Extended Data Fig. 2e-f, giving up to 61.3-fold improvement at alpha=0.006; affects every correction mask.
  • Random pattern sparsity = 0.4 (40% of superpixels ON)
    Adopted from the prior 2P-FOCUS work; controls the probe diversity and the conditioning of the inverse problem.
  • Number of random patterns per mask = 3000 (20-30% of full sampling)
    Selected from bead experiments where intensity plateaus beyond 2,000 measurements; used for all in vivo masks.
  • Superpixel grid = 100x100 superpixels, each 8x8 DMD pixels
    Fixed based on 2P-FOCUS; determines the number of unknowns in Eq. (1).
assumptions (4)
  • ad hoc to paper Two-photon fluorescence from the target is a linear function of the binary Fourier-domain illumination pattern (Eq. 1).
    This linear model is stated without derivation; physically the two-photon signal is quadratic in the field and includes cross-term interference, which is exactly the effect the method claims to exploit.
  • domain assumption The scattering medium is static during the correction and imaging window (about 1.5 minutes per plane).
    The paper states no motion artifacts were observed, but does not quantify speckle decorrelation times for deep brain or skull in vivo, which is critical for mask validity.
  • domain assumption Local intensity peaks in uncorrected images correspond to real fluorescent structures.
    Content-aware subregions choose the brightest pixel per patch as the correction target (Methods); if this peak is noise, the mask may amplify noise and bias the improvement metric.
  • domain assumption Amplitude-only modulation can select in-phase multiply scattered beams without phase control.
    The DMD blocks or passes beams but does not adjust phase; the paper argues constructive interference can be enhanced by blocking out-of-phase beams, but no derivation shows sufficient control with binary amplitude-only masks.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Compressive Fourier-Domain Intensity Coupling (C-FOCUS) enables near-millimeter deep imaging in the intact mouse brain in vivo." pith.science (2026). https://pith.science/paper/DPVCFERQ

@misc{pith2026250521822,
  author       = {Pith},
  title        = {Pith review of: Compressive Fourier-Domain Intensity Coupling (C-FOCUS) enables near-millimeter deep imaging in the intact mouse brain in vivo},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DPVCFERQ}},
  note         = {Machine review of arXiv:2505.21822}
}
read the original abstract

Two-photon microscopy is a powerful tool for in vivo imaging, but its imaging depth is typically limited to a few hundred microns due to tissue scattering, even with existing scattering correction techniques. Moreover, most active scattering correction methods are restricted to small regions by the optical memory effect. Here, we introduce compressive Fourier-domain intensity coupling for scattering correction (C-FOCUS), an active scattering correction approach that integrates Fourier-domain intensity modulation with compressive sensing for two-photon microscopy. Using C-FOCUS, we demonstrate high-resolution imaging of YFP-labeled neurons and FITC-labeled blood vessels at depths exceeding 900 um in the intact mouse brain in vivo. Furthermore, we achieve transcranial imaging of YFP-labeled dendritic structures through the intact adult mouse skull. C-FOCUS enables high-contrast fluorescence imaging at depths previously inaccessible using two-photon microscopy with 1035 nm excitation, enhancing fluorescence intensity by over 20-fold compared to uncorrected imaging. C-FOCUS provides a broadly applicable strategy for rapid, deep-tissue optical imaging in vivo.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

56 extracted references · 56 canonical work pages

  1. [1]

    Helmchen, F., Denk, W.: Deep tissue two-photon microscopy. Nat. Methods2(12), 932–940 (2005)

    work page 2005

  2. [2]

    Ji, N.: Adaptive optical fluorescence microscopy. Nat. Methods 14(4), 374–380 (2017)

    work page 2017

  3. [3]

    Rodr ´ ıguez, C., Chen, A., Rivera, J.A., Mohr, M.A., Liang, Y., Natan, R.G., Sun, W., Milkie, D.E., Bifano, T.G., Chen, X., Ji, N.: An adaptive optics module for deep tissue multiphoton imaging in vivo. Nat. Methods18(10), 1259–1264 (2021)

    work page 2021

  4. [4]

    Photonics Res

    Chen, C., Qin, Z., He, S., Liu, S., Lau, S.-F., Wu, W., Zhu, D., Ip, N.Y., Qu, J.Y.: High-resolution two-photon transcranial imaging of brain using direct wavefront sensing. Photonics Res. 9(6), 1144 (2021)

    work page 2021

  5. [5]

    Drew, P.J., Shih, A.Y., Driscoll, J.D., Knutsen, P.M., Blinder, P., Davalos, D., Akassoglou, K., Tsai, P.S., Kleinfeld, D.: Chronic optical access through a polished and reinforced thinned skull. Nat. Methods 7(12), 981–984 (2010)

    work page 2010

  6. [6]

    Papadopoulos, I.N., Jouhanneau, J.-S., Poulet, J.F.A., Judkewitz, B.: Scatter- ing compensation by focus scanning holographic aberration probing (F-SHARP). Nat. Photonics 11(2), 116–123 (2016) 19

    work page 2016

  7. [7]

    eNeuro 7(1), 0255–192019 (2020)

    Takasaki, K., Abbasi-Asl, R., Waters, J.: Superficial bound of the depth limit of two-photon imaging in mouse brain. eNeuro 7(1), 0255–192019 (2020)

    work page 2020

  8. [8]

    Mittmann, W., Wallace, D.J., Czubayko, U., Herb, J.T., Schaefer, A.T., Looger, L.L., Denk, W., Kerr, J.N.D.: Two-photon calcium imaging of evoked activity from L5 somatosensory neurons in vivo. Nat. Neurosci. 14(8), 1089–1093 (2011)

    work page 2011

Show all 56 references
  1. [9]

    Cha, J.W., Ballesta, J., So, P.T.C.: Shack-hartmann wavefront-sensor-based adaptive optics system for multiphoton microscopy. J. Biomed. Opt. 15(4), 046022 (2010)

    2010

  2. [10]

    Aviles-Espinosa, R., Andilla, J., Porcar-Guezenec, R., Olarte, O.E., Nieto, M., Levecq, X., Artigas, D., Loza-Alvarez, P.: Measurement and correction of in vivo sample aberrations employing a nonlinear guide-star in two-photon excited fluorescence microscopy. Biomed. Opt. Expr...

    2011

  3. [11]

    Wang, K., Milkie, D.E., Saxena, A., Engerer, P., Misgeld, T., Bronner, M.E., Mumm, J., Betzig, E.: Rapid adaptive optical recovery of optimal resolution over large volumes. Nat. Methods 11(6), 625–628 (2014)

    2014

  4. [12]

    Wang, K., Sun, W., Richie, C.T., Harvey, B.K., Betzig, E., Ji, N.: Direct wavefront sensing for high-resolution in vivo imaging in scattering tissue. Nat. Commun. 6, 7276 (2015)

    2015

  5. [13]

    Liu, R., Li, Z., Marvin, J.S., Kleinfeld, D.: Direct wavefront sensing enables func- tional imaging of infragranular axons and spines. Nat. Methods 16(7), 615–618 (2019)

    2019

  6. [14]

    Albert, O., Sherman, L., Mourou, G., Norris, T.B., Vdovin, G.: Smart micro- scope: an adaptive optics learning system for aberration correction in multiphoton confocal microscopy. Opt. Lett. 25(1), 52–54 (2000)

    2000

  7. [15]

    Marsh, P., Burns, D., Girkin, J.: Practical implementation of adaptive optics in multiphoton microscopy. Opt. Express 11(10), 1123–1130 (2003)

    2003

  8. [16]

    D´ ebarre, D., Botcherby, E.J., Watanabe, T., Srinivas, S., Booth, M.J., Wilson, T.: Image-based adaptive optics for two-photon microscopy. Opt. Lett. 34(16), 2495–2497 (2009)

    2009

  9. [17]

    Rueckel, M., Mack-Bucher, J.A., Denk, W.: Adaptive wavefront correction in two- photon microscopy using coherence-gated wavefront sensing. Proc. Natl. Acad. Sci. U. S. A. 103(46), 17137–17142 (2006)

    2006

  10. [18]

    Tang, J., Germain, R.N., Cui, M.: Superpenetration optical microscopy by itera- tive multiphoton adaptive compensation technique. Proc. Natl. Acad. Sci. U. S. A. 109(22), 8434–8439 (2012) 20

    2012

  11. [19]

    May, M.A., Barr´ e, N., Kummer, K.K., Kress, M., Ritsch-Marte, M., Jesacher, A.: Fast holographic scattering compensation for deep tissue biological imaging. Nat. Commun. 12(1), 4340 (2021)

    2021

  12. [20]

    Qin, Z., She, Z., Chen, C., Wu, W., Lau, J.K.Y., Ip, N.Y., Qu, J.Y.: Deep tissue multi-photon imaging using adaptive optics with direct focus sensing and shaping. Nat. Biotechnol. 40(11), 1663–1671 (2022)

    2022

  13. [21]

    I, Rosenbluh, M., Feng, S.: Memory effects in propagation of optical waves through disordered media

    Freund, I. I, Rosenbluh, M., Feng, S.: Memory effects in propagation of optical waves through disordered media. Phys. Rev. Lett. 61(20), 2328–2331 (1988)

    1988

  14. [22]

    Park, J.-H., Kong, L., Zhou, Y., Cui, M.: Large-field-of-view imaging by multi- pupil adaptive optics. Nat. Methods 14(6), 581–583 (2017)

    2017

  15. [23]

    May, M.A., Kummer, K.K., Edenhofer, M.-L., Choconta, J.L., Kress, M., Ritsch- Marte, M., Jesacher, A.: Simultaneous scattering compensation at multiple points in multi-photon microscopy. Biomed. Opt. Express 12(12), 7377–7387 (2021)

    2021

  16. [24]

    Blochet, B., Akemann, W., Gigan, S., Bourdieu, L.: Fast wavefront shaping for two-photon brain imaging with multipatch correction. Proc. Natl. Acad. Sci. U. S. A. 120(51), 2305593120 (2023)

    2023

  17. [25]

    Ren, M., Chen, J., Chen, D., Chen, S.-C.: Aberration-free 3D imaging via DMD- based two-photon microscopy and sensorless adaptive optics. Opt. Lett. 45(9), 2656–2659 (2020)

    2020

  18. [26]

    Chen, D., Ren, M., Zhang, D., Chen, J., Gu, S., Chen, S.-C.: Design of a multi-modality DMD-based two-photon microscope system. Opt. Express 28(20), 30187–30198 (2020)

    2020

  19. [27]

    Photonics Res

    Zepeda, D., Li, Y., Xue, Y.: Scattering correction through fourier-domain inten- sity coupling in two-photon microscopy (2P-FOCUS). Photonics Res. 13(4), 845 (2025)

    2025

  20. [28]

    Yoon, S., Lee, H., Hong, J.H., Lim, Y.-S., Choi, W.: Laser scanning reflection- matrix microscopy for aberration-free imaging through intact mouse skull. Nat. Commun. 11(1), 5721 (2020)

    2020

  21. [29]

    Elife 9 (2020)

    Wang, T., Wu, C., Ouzounov, D.G., Gu, W., Xia, F., Kim, M., Yang, X., Warden, M.R., Xu, C.: Quantitative analysis of 1300-nm three-photon calcium imaging in the mouse brain. Elife 9 (2020)

    2020

  22. [30]

    Tung, C.-K., Sun, Y., Lo, W., Lin, S.-J., Jee, S.-H., Dong, C.-Y.: Effects of objec- tive numerical apertures on achievable imaging depths in multiphoton microscopy. Microsc. Res. Tech. 65(6), 308–314 (2004)

    2004

  23. [31]

    Elife 6 (2017)

    Kondo, M., Kobayashi, K., Ohkura, M., Nakai, J., Matsuzaki, M.: Two-photon 21 calcium imaging of the medial prefrontal cortex and hippocampus without cortical invasion. Elife 6 (2017)

    2017

  24. [32]

    Wang, T., Ouzounov, D.G., Wu, C., Horton, N.G., Zhang, B., Wu, C.-H., Zhang, Y., Schnitzer, M.J., Xu, C.: Three-photon imaging of mouse brain structure and function through the intact skull. Nat. Methods 15(10), 789–792 (2018)

    2018

  25. [33]

    Sohmen, M., Borozdova, M., Ritsch-Marte, M., Jesacher, A.: Complex-valued scatter compensation in nonlinear microscopy. Phys. Rev. Appl. 22(4), 044036 (2024)

    2024

  26. [34]

    Elife 11, 73266 (2022)

    Xue, Y., Waller, L., Adesnik, H., P´ egard, N.: Three-dimensional multi-site random access photostimulation (3D-MAP). Elife 11, 73266 (2022)

    2022

  27. [35]

    Belthangady, C., Royer, L.A.: Applications, promises, and pitfalls of deep learning for fluorescence image reconstruction. Nat. Methods 16(12), 1215–1225 (2019)

    2019

  28. [36]

    Xue, Y., Boivin, J.R., Wadduwage, D.N., Park, J.K., Nedivi, E., So, P.T.C.: Mul- tiline orthogonal scanning temporal focusing (mosTF) microscopy for scattering reduction in in vivo brain imaging. Sci. Rep. 14(1), 10954 (2024)

    2024

  29. [37]

    Sci Adv 7(28) (2021)

    Zheng, C., Park, J.K., Yildirim, M., Boivin, J.R., Xue, Y., Sur, M., So, P.T.C., Wadduwage, D.N.: De-scattering with excitation patterning enables rapid wide- field imaging through scattering media. Sci Adv 7(28) (2021)

    2021

  30. [38]

    Xue, Y., Berry, K.P., Boivin, J.R., Wadduwage, D., Nedivi, E., So, P.T.C.: Scattering reduction by structured light illumination in line-scanning temporal focusing microscopy. Biomed. Opt. Express, BOE 9(11), 5654–5666 (2018)

    2018

  31. [39]

    Res Sq (2023)

    Wei, Z., Boivin, J.R., Xue, Y., Burnell, K., Wijethilake, N., Chen, X., So, P.T.C., Nedivi, E., Wadduwage, D.N.: De-scattering deep neural network enables fast imaging of spines through scattering media by temporal focusing microscopy. Res Sq (2023)

    2023

  32. [40]

    Xue, Y., Ren, D., Waller, L.: Three-dimensional bi-functional refractive index and fluorescence microscopy (BRIEF). Biomed. Opt. Express (2022)

    2022

  33. [41]

    Li, X., Zhang, G., Wu, J., Zhang, Y., Zhao, Z., Lin, X., Qiao, H., Xie, H., Wang, H., Fang, L., Dai, Q.: Reinforcing neuron extraction and spike inference in calcium imaging using deep self-supervised denoising. Nat. Methods 18(11), 1395–1400 (2021)

    2021

  34. [42]

    Qu, L., Zhao, S., Huang, Y., Ye, X., Wang, K., Liu, Y., Liu, X., Mao, H., Hu, G., Chen, W., Guo, C., He, J., Tan, J., Li, H., Chen, L., Zhao, W.: Self-inspired learning for denoising live-cell super-resolution microscopy. Nat. Methods 21(10), 1895–1908 (2024) 22

    2024

  35. [43]

    Qiao, C., Zeng, Y., Meng, Q., Chen, X., Chen, H., Jiang, T., Wei, R., Guo, J., Fu, W., Lu, H., Li, D., Wang, Y., Qiao, H., Wu, J., Li, D., Dai, Q.: Zero-shot learning enables instant denoising and super-resolution in optical fluorescence microscopy. Nat. Commun. 15(1), 4180 (2024)

    2024

  36. [44]

    Cao, R., Li, Y., Zhou, Y., Li, M., Lin, F., Wang, W., Zhang, G., Wang, G., Jin, B., Ren, W., Sun, Y., Zhao, Z., Zhang, W., Sun, J., Hou, Y., Xu, X., Hu, J., Shi, W., Fu, S., Liang, Q., Lu, Y., Li, C., Zhao, Y., Li, Y., Kuang, D., Wu, J., Fei, P., Qu, J., Xi, P.: Dark-based opt...

    2025

  37. [45]

    Qureshi, M.M., Brake, J., Jeon, H.-J., Ruan, H., Liu, Y., Safi, A.M., Eom, T.J., Yang, C., Chung, E.: In vivo study of optical speckle decorrelation time across depths in the mouse brain. Biomed. Opt. Express 8(11), 4855–4864 (2017)

    2017

  38. [46]

    Cheng, H., Tong, S., Deng, X., Liu, H., Du, Y., He, C., Qiu, P., Wang, K.: Deep- brain 2-photon fluorescence microscopy in vivo excited at the 1700 nm window. Opt. Lett. 44(17), 4432–4435 (2019)

    2019

  39. [47]

    Optica 7(8), 947–960 (2020) https://doi.org/10.1364/OPTICA.395825

    Wang, T., Xu, C.: Three-photon neuronal imaging in deep mouse brain. Optica 7(8), 947–960 (2020) https://doi.org/10.1364/OPTICA.395825

    2020 doi

  40. [48]

    Ouzounov, D.G., Wang, T., Wang, M., Feng, D.D., Horton, N.G., Cruz- Hern´ andez, J.C., Cheng, Y.-T., Reimer, J., Tolias, A.S., Nishimura, N., Xu, C.: In vivo three-photon imaging of activity of GCaMP6-labeled neurons deep in intact mouse brain. Nat. Methods 14(4), 388–390 (2017)

    2017

  41. [49]

    Streich, L., Boffi, J.C., Wang, L., Alhalaseh, K., Barbieri, M., Rehm, R., Deivasigamani, S., Gross, C.T., Agarwal, A., Prevedel, R.: High-resolution struc- tural and functional deep brain imaging using adaptive optics three-photon microscopy. Nat. Methods 18(10), 1253–1258 (2021)

    2021

  42. [50]

    Packer, A.M., Peterka, D.S., Hirtz, J.J., Prakash, R., Deisseroth, K., Yuste, R.: Two-photon optogenetics of dendritic spines and neural circuits. Nat. Methods 9(12), 1202–1205 (2012)

    2012

  43. [51]

    Prakash, R., Yizhar, O., Grewe, B., Ramakrishnan, C., Wang, N., Goshen, I., Packer, A.M., Peterka, D.S., Yuste, R., Schnitzer, M.J., Deisseroth, K.: Two-photon optogenetic toolbox for fast inhibition, excitation and bistable modulation. Nat. Methods 9(12), 1171–1179 (2012)

    2012

  44. [52]

    Zhang, Z., Russell, L.E., Packer, A.M., Gauld, O.M., H¨ ausser, M.: Closed-loop all- optical interrogation of neural circuits in vivo. Nat. Methods 15(12), 1037–1040 (2018)

    2018

  45. [53]

    Adesnik, H., Abdeladim, L.: Probing neural codes with two-photon holographic optogenetics. Nat. Neurosci. 24(10), 1356–1366 (2021) 23

    2021

  46. [54]

    Nature Communications 12(1), 2411 (2021)

    Ruan, H., Xu, J., Yang, C.: Optical information transmission through com- plex scattering media with optical-channel-based intensity streaming. Nature Communications 12(1), 2411 (2021)

    2021

  47. [55]

    Optica, OPTICA 4(10), 1306–1313 (2017)

    Zhang, J., P´ egard, N., Zhong, J., Adesnik, H., Waller, L.: 3D computer-generated holography by non-convex optimization. Optica, OPTICA 4(10), 1306–1313 (2017)

    2017

  48. [56]

    Beck, A., Teboulle, M.: A fast iterative shrinkage-thresholding algorithm for linear inverse problems. SIAM J. Imaging Sci. 2(1), 183–202 (2009) Appendix A Extended Data Fig.1 Appendix B Extended Data Fig.2 Appendix C Extended Data Fig.3 24 g Ex vivo mouse skull h In vivo mous...

    2009

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.