Ultrahigh-speed digital holography reveals blood flow beneath the skin

arXiv:2611.15535 · physics.optics · Submitted 2026-11-11 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Ultrahigh-speed digital holography reveals blood flow beneath the skin".

Mira: Here's the gist of the paper. Laser Doppler holography (LDH) leverages ultrahigh-speed digital holography to reveal bloodflow signals in vessels beneath the skin of the closed eyelid and around the eye This approach demonstrates that branching vascular patterns can be revealed beneath the skin without requiring elaborate immobilization procedures The on-axis LDH prototype used for eyelid and periocular imaging is shown in Fig.

Kai: 2 The optical configuration derives from an inline Mach–Zehnder digital holography ophthalmoscope operating at λ0 ∼ 852 nm Light from a diode laser (Thorlabs FPV852P) was divided between the illumination and reference arms in a 90:10 power ratio using polarizationmaintaining fibers The illumination arm contained an engineered diffuser (Thorlabs ED1-C20-MD, 20◦ circular top-hat) and three biconvex relay lenses with focal lengths of 60, 60, and 25 mm Cross-polarized backscattered light from the imaged region was recombined with the reference field and recorded interferometrically with a Phantom S711 camera (Ametek Vision Research) over 768 × 512 pixels with a d = 20 µm pixel pitch Data Acquisition and Processing Recording conditions involved imaging both eyes of a healthy 25-year-old man with the lids closed and open after written informed consent Each recording comprised 65,536 interferograms acquired at approximately 16,000 frames/s (approximately 62 µs frame period), with a 58 µs exposure, for a duration of approximately 4 s Fresnel propagation used a numerical reconstruction distance of 508 mm Principal-component analysis (PCA)

49: used 32-frame windows with a stride of 16 frames Components associated with the 8 smallest singular values were retained for the red channel, and those associated with the 28 smallest singular values for the cyan channel Sixteen successive PCA outputs were averaged in a rolling buffer before color-channel intensity mapping Key Findings The closed-eye views reveal distinct branching vascular patterns across both upper lids Over the right lid, a prominent central branch extends upward from near the lash line and divides toward the upper field, with curved branches extending toward the inner corner With the eyes open (Fig.

Mira: 1c,d), vascular contrast remains visible in the skin above and around the eyes A curved vessel is prominent above the right upper lid, with further contrast toward the inner canthus and lower periocular skin Both openeye views contain a bright return within the pupil, spatially distinct from the periocular vascular field Clinical Implications and Limitations Periocular vascular monitoring could complement surface inspection during reconstruction and inflammatory disease follow-up Existing optical studies have monitored eyelid graft revascularization, flap perfusion, and responses to local anesthetics Limitations include the lack of absolute flow quantification, depth localization, or evidence of treatment response in this single-participant demonstration Numerical refocusing does not establish vessel depth; OCT or photoacoustic imaging could provide complementary localization Skin displacement and eye movement can contribute to coherent-light fluctuations LDH reveals branching vascular contrast beneath the closed eyelid and in the skin surrounding the open eye, enabling blood-flow visualization beneath the skin without requiring elaborate immobilization procedures The openly available recordings support further investigation of blood-flow dynamics beneath the skin and their potential value for vascular monitoring REFERENCES

47: M.

Kai: Dubosc, M.

Paper summary: Kai: So, wrapping up on this paper, "Ultrahigh-speed digital holography reveals blood flow beneath the skin," they successfully demonstrated that you can reveal branching vascular contrast beneath the closed eyelid and in the skin surrounding an open eye <ref:2611.15535#pg3>.

Mira: The authors are showing that this approach lets you see blood flow signals without needing any elaborate immobilization procedures, which is a big deal for practical applications <ref:2611.15535#pg1>.

Lev: The implication for us in hardware development is that we need to focus on how those numerical reconstruction and filtering techniques actually translate to stable real-world measurements when you are dealing with moving biological tissue <ref:2611.15535#pg2>.

Kai: And they're pointing toward the future, suggesting that these openly available recordings support further investigation into blood flow dynamics beneath the skin and their potential value for vascular monitoring <ref:2611.15535#pg3>.

Mira: It’s a solid piece of work because it connects surface appearance with underlying vascular contrast in a way that was hard to see before <ref:2611.15535#pg3>.

Lev: But remember the caveat they put out, which is that numerical refocusing doesn't establish vessel depth, and you’d need things like OCT or photoacoustic imaging for that localization sixty-one <ref:2611.15535#pg2>.

Kai: So to summarize this paper, it shows us a way to visualize blood flow beneath the skin using high-speed digital holography on the eyelid and around the eye <ref:2611.15535#pg1>.

Mira: It’s about proving that you can see those branching vascular patterns without needing to fix the patient in place <ref:2611.15535#pg3>.

Lev: And the next step is figuring out how to get those measurements stable enough for actual use, given all the potential fluctuations mentioned in the paper <ref:2611.15535#pg2>.

Conclusion: Kai: So we're wrapping up this look at "Ultrahigh-speed digital holography reveals blood flow beneath the skin." The core idea is showing that you can see branching vessel patterns right under the eyelid and around the eye without needing to keep anyone still.

Mira: Yeah, it’s about using high speed digital holography to get those signals. It means they're detecting blood flow in vessels beneath the skin, and they found that you can see how those vessels branch out even when the person's eyes are closed.

Lev: From an error correction standpoint, seeing a pattern like that is interesting because it shows you what kind of signal quality you’re dealing with in real-world biological tissue. It’s not perfect data, but it gives us a baseline for what we can expect from these kinds of imaging systems.

Kai: Right, the authors are using an on-axis prototype that's pretty complex optically—a Mach–Zehnder setup operating at eight hundred fifty two nanometers light—to get those images. It’s about showing what’s actually built and measured, not just a simulation.

Mira: And the authors did this using light from a diode laser, splitting it up for illumination and reference arms with a ninety to ten power ratio using polarization maintaining fibers. That gets you that cross-polarized backscattered light needed for the digital holography part.

Lev: The data acquisition itself is intense, acquiring sixty-five thousand five hundred thirty-six interferograms at about sixteen thousand frames per second, which means a frame period of roughly sixty two microseconds. That’s a lot of data to process and store before you even get to the analysis part.

Kai: And the processing involved using principal component analysis on thirty two frame windows with a stride of sixteen frames. They kept components associated with the eight smallest singular values for the red channel and twenty eight smallest for the cyan channel. That’s how they extracted those vascular features from all that raw noise.

Mira: But what's important is what they found in the results—the closed-eye views clearly show distinct branching vascular patterns across both upper lids, with a prominent central branch extending upward on the right lid and curved branches going toward the inner corner. It’s visual evidence of blood flow under the skin that wasn't obvious before.

Lev: I agree, those visual findings are compelling because they demonstrate that the signal is present in these subcutaneous vessels, which is what we’re aiming for when we think about monitoring things like flap perfusion later on. But remember, they also noted a limitation there—numerical refocusing doesn't actually tell you the depth of those vessels.

Kai: Exactly. So the takeaway here is that LDH can reveal this branching vascular contrast beneath the closed eyelid and in the skin around an open eye without needing any immobilization procedures. It opens up a new window for visualizing blood flow right under our skin.

Mira: This work supports further investigation into blood flow dynamics beneath the skin, which could have real value for monitoring things like wound healing or even certain types of inflammatory diseases. It’s a strong piece of evidence showing the potential of this imaging modality.

Lev: The limitation they highlighted is that you still need complementary techniques like OCT or photoacoustic imaging to properly localize those vessels in terms of depth, which is a necessary caveat for moving from demonstration to clinical monitoring tools.

Kai: So we see the potential for seeing these patterns without fixing the patient, and we know what’s missing—the precise depth information. That leads us right into how error correction and hardware stability can actually make this kind of measurement reliable enough to be useful in a real medical setting next.

Institut Langevin · CNRS · ESPCI Paris · PSL University

physics.optics

Submitted: 2026-11-11

Updated: 2026-11-11

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

Importance score: 82/100

The gist: Laser Doppler holography (LDH) leverages ultrahigh-speed digital holography to reveal bloodflow signals in vessels beneath the skin of the closed eyelid and around the eye This approach demonstrates…

Key concepts

Laser Doppler Holography (LDH)
LDH combines ultrahigh-speed digital holography with laser Doppler techniques. It captures light signals from blood flow within vessels under the skin of the eyelid and around the eye. This allows researchers to see how blood moves in these small vessels in real-time.
Digital Holography
Digital holography is a technique that uses digital sensors to capture interference patterns from light waves. It reconstructs a 3D image by analyzing how light reflects off different parts of an object. In this study, it was used to record and analyze the complex light signals generated by blood flow.
Principal-component analysis (PCA)
PCA is a mathematical technique used to reduce the dimensionality of data while retaining most of its important information. In this research, PCA was applied to the recorded interferograms to filter out noise and isolate the specific components of vascular signals associated with different color channels.
Vascular Branching Patterns
These are the distinct networks where blood vessels split into smaller vessels. The study found clear examples of these branching patterns beneath both upper eyelids, showing how blood flows in complex, interconnected routes within the subcutaneous tissue.

Terminology

Summary

Laser Doppler holography (LDH) leverages ultrahigh-speed digital holography to reveal bloodflow signals in vessels beneath the skin of the closed eyelid and around the eye This approach demonstrates that branching vascular patterns can be revealed beneath the skin without requiring elaborate immobilization procedures

Methodology

The on-axis LDH prototype used for eyelid and periocular imaging is shown in Fig. 2 The optical configuration derives from an inline Mach–Zehnder digital holography ophthalmoscope operating at λ0 ∼ 852 nm Light from a diode laser (Thorlabs FPV852P) was divided between the illumination and reference arms in a 90:10 power ratio using polarizationmaintaining fibers The illumination arm contained an engineered diffuser (Thorlabs ED1-C20-MD, 20◦ circular top-hat) and three biconvex relay lenses with focal lengths of 60, 60, and 25 mm Cross-polarized backscattered light from the imaged region was recombined with the reference field and recorded interferometrically with a Phantom S711 camera (Ametek Vision Research) over 768 × 512 pixels with a d = 20 µm pixel pitch

Data Acquisition and Processing

Recording conditions involved imaging both eyes of a healthy 25-year-old man with the lids closed and open after written informed consent Each recording comprised 65,536 interferograms acquired at approximately 16,000 frames/s (approximately 62 µs frame period), with a 58 µs exposure, for a duration of approximately 4 s Fresnel propagation used a numerical reconstruction distance of 508 mm Principal-component analysis (PCA) [49] used 32-frame windows with a stride of 16 frames Components associated with the 8 smallest singular values were retained for the red channel, and those associated with the 28 smallest singular values for the cyan channel Sixteen successive PCA outputs were averaged in a rolling buffer before color-channel intensity mapping

Key Findings

The closed-eye views reveal distinct branching vascular patterns across both upper lids Over the right lid, a prominent central branch extends upward from near the lash line and divides toward the upper field, with curved branches extending toward the inner corner With the eyes open (Fig. 1c,d), vascular contrast remains visible in the skin above and around the eyes A curved vessel is prominent above the right upper lid, with further contrast toward the inner canthus and lower periocular skin Both openeye views contain a bright return within the pupil, spatially distinct from the periocular vascular field

Clinical Implications and Limitations

Periocular vascular monitoring could complement surface inspection during reconstruction and inflammatory disease follow-up Existing optical studies have monitored eyelid graft revascularization, flap perfusion, and responses to local anesthetics Limitations include the lack of absolute flow quantification, depth localization, or evidence of treatment response in this single-participant demonstration Numerical refocusing does not establish vessel depth; OCT or photoacoustic imaging could provide complementary localization Skin displacement and eye movement can contribute to coherent-light fluctuations

Conclusion

LDH reveals branching vascular contrast beneath the closed eyelid and in the skin surrounding the open eye, enabling blood-flow visualization beneath the skin without requiring elaborate immobilization procedures The openly available recordings support further investigation of blood-flow dynamics beneath the skin and their potential value for vascular monitoring

REFERENCES

[47] M. Dubosc, M. Boy-Arnould, J. p. 0

Where the paper says it

Ultrahigh-speed digital holography reveals blood flow beneath the skin Yann Fischer,1 Zacharie Auray,1 Maxime Boy-Arnould,1 and Michael Atlan1 1 Institut Langevin, CNRS, ESPCI Paris, PSL University, Paris, France (Dated: 11 November 2026) Laser Doppler holography (LDH) leverages ultrahigh-speed digital holography to reveal bloodflow signals in vessels beneath the skin of the closed eyelid and around the eye. We demonstrate this in one volunteer using recordings acquired at approximately 16,000 frames/s under 10 mW of diffuse near-infrared illumination.

Guillou, et al. p. 3
Where the paper says it

13347v1 [physics. Guillou, B.

, Holovibes: real-time ultrahigh-speed digital hologram rendering and short-time analysis (2025), preprint, arXiv:2508. p. 3
Where the paper says it

Guillou, et al. , Holovibes: real-time ultrahigh-speed digital hologram rendering and short-time analysis (2025), preprint, arXiv:2508.

03911 [physics. p. 3
Where the paper says it

, Holovibes: real-time ultrahigh-speed digital hologram rendering and short-time analysis (2025), preprint, arXiv:2508. 03911 [physics.

optics] 11 Nov 2026

[48] Holovibes, Software, version 14.14.1

[5] J. D. Briers and S. p. 2

Where the paper says it

Express 2, 1470 (2011). Briers and S.

Webster, J. p. 2
Where the paper says it

Briers and S. Webster, J.

Biomed Opt. p. 2
Where the paper says it

Strömberg, and I. Fredriksson, J Biomed Opt 25, 1 (2020).

1, 174 (1996)

[6] J. D. Briers, J. Opt Soc Am A 13, 345 (1996)

[7] A B Parthasarathy, W J Tom, A Gopal, X Zhang, and A K Dunn, Opt Express 16, 1975 (2008)

[8] D A Rivera and C B Schaffer Biomed. Opt Express 14, 3950 (2023)

[9] M Hultman, I Fredriksson, M Larsson, A Alvandpour, and T Strömberg J Biophotonics 11, e201700069 (2018)

[10] M Hultman, M Larsson, T Strömberg, and I Fredriksson J Biomed Opt 25, 1 (2020)

[11] E Colin, A Plyer, M Golzio, N Meyer G Favre and X Orlik J Biomed Opt 27, 046003 (2022)

[12] G Mansutti, M Villiger B E Bouma and N UribePatarroyo Biomed. Opt Express 15, 5756 (2024)

[13] C Puissant, P Abraham, S Durand, A HumeauHeurtier S Faure G Lefthériotis P Rousseau and G Mahé PLoS One 8, e61320 (2013)

[14] M Roustit, C Millet, S Blaise B Dufournet and J Cracowski Microvasc Res 80, 505 (2010)

[15] B Ruaro, A Sulli E Alessandri C Pizzorni G Ferrari and M Cutolo Ann Rheum Dis 73, 1181 (2014)

[16] C Millet, M Roustit, S Blaise and J Cracowski Microvasc Res 82, 147 (2011)

[59] M Hu, C Wang, Y Li, H Zhang H Li Q Dai H Lian Y e Zhao and Y Fu Transl Vis Sci Technol. 13, 6 (2024)

[60] Y Gao, M Huang W Dong Y Li Y Cheng F Yang H Li and X Yan Eye 40, 521 (2026)

[61] G Hoarau A L Best S Zina-Meziou M BenaliAbdallah M Loukil M Bouvet E Barreau and M Labetoulle J Ocul. Pharmacol Ther. 41, 24 (2025)

[63] J Vennström Berggren, J Nääv Ottosson K Tenland, M Naumovska M Fontana A Merdasa and M Malmsjö Orbit 45, 48 (2026)

[64] M Stridh U Dahlstrand M Naumovska K Engelsberg B Gesslein R Sheikh A Merdasa and M Malmsjö Orbit 43, 453 (2024)

[65] C Ujita, Y Hashimoto K Noguchi N Nakamura M Yoshimura S Ichiki M Uehara A Nakazaki T Taniguchi and T Yoshitomi Cureus 16, e75118 (2024)

[69] J M Bland and D G Altman Lancet 327, 307 (1986)

[51] High-speed laser Doppler holography of the eyelid, facial skin and anterior segment of the eye of a volunteer, revealing blood flow responses in subcutaneous vessels, Zenodo, raw dataset (2026)

[52] J Berggren N Castelo K Tenland U Dahlstrand K Engelsberg S Lindstedt R Sheikh and M Malmsjö Ophthalmic Plast. Reconstr. Surg. 37, 324 (2021)

[53] K Tenland J Berggren K Engelsberg E Bohman U Dahlstrand N Castelo S Lindstedt R Sheikh and M Malmsjö Ophthalmic Plast. Reconstr. Surg. 37, 168 (2021)

[54] K Tenland J V Berggren C Dybelius Ansson J Hult U Dahlstrand S Lindstedt R Sheikh and M Malmsjö Ophthalmic Plast. Reconstr. Surg. 36, 148 (2020)

[55] K Tenland M Naumovska M Stridh R Sheikh A Merdasa and J Vennström Berggren Orbit 44, 555 (2025)

[56] A H C Wiktorin J V Berggren M Malmsjö S Lindstedt R Sheikh and E Bohman Ophthalmic Plast. Reconstr. Surg. 38, 588 (2022)

[57] R Sheikh U Dahlstrand K Memarzadeh J Blohmé N Reistad and M Malmsjö Ophthalmic Plast. Reconstr. Surg. 34, 123 (2018)

[58] J Bunke A Merdasa M Stridh P Rosenquist J Berggren J E Hernandez-Palacios U Dahlstrand N Reistad R Sheikh and M Malmsjö Ophthalmic Plast. Reconstr. Surg. 38, 462 (2022)

[50] L Puyo C David R Saad S Saad J Gautier J A Sahel V Borderie M Paques and M Atlan Biomed. Opt Express 12, 4478 (2021)

[51] High-speed laser Doppler holography of the eyelid, facial skin and anterior segment of the eye of a volunteer, revealing blood flow responses in subcutaneous vessels, Zenodo, raw dataset (2026)

[53] K Tenland J Berggren K Engelsberg E Bohman U Dahlstrand N Castelo S Lindstedt R Sheikh and M Malmsjö Ophthalmic Plast. Reconstr. Surg. 37, 168 (2021)

[54] K Tenland J V Berggren C Dybelius Ansson J Hult U Dahlstrand S Lindstedt R Sheikh and M Malmsjö Ophthalmic Plast. Reconstr. Surg. 36, 148 (2020)

[55] K Tenland M Naumovska M Stridh R Sheikh A Merdasa and J Vennström Berggren Orbit 44, 555 (2025)

[66] L An J Qin and R K Wang Opt Express 18, 8220 (2010)

[67] H F Zhang K Maslov G Stoica and L V Wang Nat. Biotechnol. p. 4

Where the paper says it

5 Biotechnol. 24, 848 (2006).

24, 848 (2006)

[68] A Chizari T Knop B Sirmacek F van der Heijden and W Steenbergen Biomed. Opt. Express 11, 2352 (2020)

[69] J M Bland and D G Altman Lancet 327, 307 (1986)

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Deep learning models can be trained to perform numerical Fresnel reconstruction from ultrahigh-speed digital holography data, enabling automated visualization of branching vascular patterns in complex scattering media.

  2. The improved system can visualize blood flow signals beneath the skin without requiring elaborate immobilization procedures, extending ocular imaging systems to blood-flow visualization beneath the skin.

  3. Principal component analysis (PCA) can be used to filter and reveal vascular contrast by retaining components associated with specific singular values, allowing the system to encode different retained subspaces in red and cyan channels reflecting Doppler broadening.

Abstract

Laser Doppler holography (LDH) leverages ultrahigh-speed digital holography to reveal blood-flow signals in vessels beneath the skin of the closed eyelid and around the eye. We demonstrate this in one volunteer using recordings acquired at approximately 16,000 frames/s under 10 mW of diffuse near-infrared illumination. Numerical Fresnel reconstruction and principal-component filtering reveal branching vascular patterns. These results extend an ocular imaging system to blood-flow visualization beneath the skin without requiring elaborate immobilization procedures.

Sources

Related papers