How scattered light and precisely timed pulses reveal a thin slice of life—across a wider view.

How scattered light and precisely timed pulses reveal a thin slice of life—across a wider view.

AI-generated graphical abstract · Conceptual schematic, not an optical ray diagram.

To see a layer, make only that layer light up.

A microscope looking into thick tissue sees light from many depths at once. That unwanted glow can hide the details you care about.

This study combines three ideas: compress a laser pulse only at the chosen depth, turn its illumination into a pattern of bright speckles, and sweep that small patch across the specimen.

A short pulse makes two-photon fluorescence strongest in one thin layer.

Speckle spreads the light over the objective’s pupil, reducing the risk of lens damage.

A camera gathers light as the small patch moves across a larger field.

Find the glowing layer.

Move the focal plane through a virtual tissue sample. Watch a reference layer at 0 µm brighten as the pulse comes into focus.

Green = fluorescence · violet = selected focal plane Positions and intensity are illustrative; tissue is synthetic.

Temporal focusing

At the reference layer, the pulse is shortest and fluorescence is strongest.

A teaching model: a Gaussian response with 5 µm full width at half maximum when focusing is on. The broad 20 µm “off” curve is illustrative, not a measured comparison.

↗ Try this Move to +5 µm, then turn temporal focusing off. Notice how much glow remains outside the selected layer.

A small patch. A big picture.

Illuminating a whole field at once needs a lot of laser power. Sweep a ~50 µm patch instead, and let the camera collect the picture.

Simplified raster scan and synthetic cells. Playback is slowed down; it does not represent experimental frame timing.

Sweep across the specimen

Each illuminated region contributes to one camera exposure. The speckle texture averages out as the patch moves.

The camera is ready. Start the scan to collect light across the field.

The paper uses galvo mirrors to steer the patch. Its demonstrated bead acquisition takes 100 ms per frame; biological samples can need longer exposures.

Random light has a second job.

Keep thousands of separate speckle images, and their changing patterns can reveal finer detail. This is random illumination microscopy, or RIM.

Less background. More contrast.

In mouse intestinal villi, temporal focusing helped RIM reject glow from other depths.

Reported contrast in this sample · 5,000 frames × 10 ms = 50 s of exposure, before overhead and reconstruction.

Finer detail, with more measurements.

Reported lateral resolution improvement along the x and y directions, compared with the averaged temporally focused image.

20,000 frames × 5 ms = 100 s of exposure, before overhead and reconstruction. Improvement measured on fluorescent beads.

Two modes, different tradeoffs. Fast scanning collects a wide-field image in one exposure. Super-resolution RIM collects many separate images and reconstructs detail afterward. The paper does not demonstrate super-resolution at 10 frames per second.

Research notes

Two-photon excitation depends strongly on instantaneous light intensity. A diffraction grating separates the pulse’s colors. They recombine into a short pulse at the focal plane, where peak intensity is high. Away from that plane the pulse is stretched in time, reducing peak intensity and fluorescence. This is temporal focusing: a thin illuminated layer, rather than simply a tiny point.

In conventional temporal focusing, light can be concentrated along one thin spectral line at the objective’s back pupil. Speckle illumination distributes energy over many spectral lines, lowering local power density and the risk of objective damage. The paper also reports improved optical sectioning in scattering media. In the separate RIM mode, translating the speckles provides intensity variations that a reconstruction algorithm uses to retrieve detail.

The ~5 µm axial width was measured at 800 nm with a 20×, NA 0.75 objective. The abstract reports a 300 × 300 µm field; detailed bead results use 330 × 330 µm. The maximum 10 frames/s result is for fluorescent beads. Cells, larval epidermis and organoids were also imaged at different, generally longer exposure times. The two interactive drawings on this page use synthetic tissue and simplified models, not the paper’s microscopy data. The opening image is an AI-generated conceptual illustration.

By combining temporal focusing, speckle illumination and scanning, the researchers address several limitations of camera-based two-photon microscopy at once: section thickness, field size and concentrated power at the objective. Performance still depends on the sample, scattering, laser settings and acquisition mode.

An interactive interpretation of an open-access research paper.

Speckle illumination temporal focusing two-photon excited fluorescence microscopy

Federico Vernuccio, Michał Marynowski, Xiangyi Li, Assia Benachir, Luca Genchi, Randy Bartels, Sandro Heuke, Anne Sentenac & Hervé Rigneault. Optica 13, 577–586 (2026). DOI: 10.1364/OPTICA.581520.

Source article published under CC BY 4.0. Original explanatory illustrations and interactive models on this page.