With random lasers, Yale researchers fight random noise, improve imaging

Apr 30, 2012
With random lasers, Yale researchers fight random noise, improve imaging
Light emerges from a random laser. (Image byBrandon Redding, Yale)

(Phys.org) -- Using “random lasers” as a source of illumination in medical imaging equipment could improve both processing time and the clarity of the final images, according to new research by Yale University scientists.

Imaging systems currently rely on a variety of light sources — specialty light bulbs, light-emitting diodes (LEDs), and traditional lasers. But systems using traditional lasers, the brightest of these light sources, often yield undesirable visual byproducts that mar the final picture. One common byproduct, speckle, looks something like a snowfall pattern.

The Yale researchers have engineered a special type of called a random laser — which generates and emits light differently from traditional lasers — to serve the same function without giving off speckle or other visual blight. They report their results online April 29 in the journal Nature Photonics.

“Our work is innovative and significant because we show that random lasers are much brighter than LEDs and light bulbs and also generate speckle-free images,” said Michael A. Choma, an assistant professor of diagnostic radiology, pediatrics, and biomedical engineering at Yale. He is one of the study’s principal investigators.

Hui Cao, a professor of applied physics and physics at Yale, is the other principal investigator. Brandon Redding, a postdoctoral associate in applied physics, is the lead author.

A traditional laser emits a single intense beam of light, known as a spatial mode. Photons from that single beam can be scattered by a sample under observation, resulting in random grainy background noise — speckle — on top of the desired image.

One way of mitigating the noise is to use many different spatial modes, such as the light emitted by a LED or light bulb. Unfortunately, these light sources are dim compared with lasers.

But random lasers offer the best of both worlds, according to the Yale researchers. They are bright, like lasers, while also having many modes, like a light bulb, so they generate speckle-free images. That is, random lasers are something akin to a with the intensity of a laser.

“Our random lasers combine the advantages of lasers and the white light sources, and may be used for a wide range of imaging and projection applications,” said Cao.

The light emitted by random lasers could also enable faster image generation. This would help researchers and clinicians better capture fast-moving physiological phenomena — the movements of embryo hearts, perhaps, or blood flow patterns in the eye — as well as broad swaths of tissue in less time than required by current technologies.

“Your light source really defines the boundaries of what you can do — how fast you can image,” said Choma. “And you always want to go faster.”

Random lasers could have applications in consumer electronics also, according to the researchers — in digital projection systems, for example.

Within medical imaging, the introduction of random lasers could lead to improved microscopy and endoscopy, they said.

The scientists have produced a prototype random laser for use in imaging applications and are refining it.

The National Science Foundation and the Yale Child Health Research Center supported the work.

The paper is titled “Speckle-free laser imaging using random laser illumination.”

Explore further: Experiment with speeding ions verifies relativistic time dilation to new level of precision

Related Stories

Scientists predict paradoxical laser effect

Apr 25, 2012

New laser-effect, discovered by scientists from the Vienna University of Technology, Princeton, Yale and ETH Zurich: If coupled, lasers can switch each other off, leading to a "laser blackout".

New lasing technique inspired by brightly colored birds

May 09, 2011

(PhysOrg.com) -- Researchers at Yale University have succeeded in building a new kind of laser based on the way brightly colored birds show their colors. Building on the new approach to creating laser beams, ...

Quantum leap for phonon lasers

Feb 22, 2010

Physicists have taken major step forward in the development of practical phonon lasers, which emit sound in much the same way that optical lasers emit light. The development should lead to new, high-resolution ...

Scientists show atoms act like lasers

Feb 25, 2011

(PhysOrg.com) -- Scientists from The Australian National University have developed an atom laser that behaves exactly like a light laser, opening up new possibilities in things like holograms.

Recommended for you

How Paramecium protozoa claw their way to the top

Sep 19, 2014

The ability to swim upwards – towards the sun and food supplies – is vital for many aquatic microorganisms. Exactly how they are able to differentiate between above and below in often murky waters is ...

User comments : 8

Adjust slider to filter visible comments by rank

Display comments: newest first

antialias_physorg
5 / 5 (1) Apr 30, 2012
Pretty neat for a lighting application. But are they not just averaging out one kind of noise by adding another kind of noise (superposition of speckle from different modes)?

The question being: Since this does not seem to give a more pure form of laserlight the information content gathered from imaging systems is probably not improved.
Eikka
not rated yet Apr 30, 2012
The question being: Since this does not seem to give a more pure form of laserlight the information content gathered from imaging systems is probably not improved.


Getting a pure laser light is not actually better in imaging unless you're interested in very particular features, because it interacts with the sample only in certain ways. It has only one frequency and one phase, and one spatial distribution, which creates interference patterns.

Shuffling the phase, frequency and spatial distribution around is equivalent to sampling at multiple frequencies and at different times and locations, so the noise is overwhelmed by correct information from all the other samples when you average it out.

It's like placing a grid in front of your eyes which blocks some of your visibility, but if you start to vibrate the grid in all directions, suddenly it seems to dissapear and you can see through.
antialias_physorg
5 / 5 (1) Apr 30, 2012
That#s the trick then, though. Different frequencies interact dfferently (i.e. give you information about different structures to different degrees). The same 'problem' occurs in dual source CT wih differing energies.
You get information about more structures (as could be expected from hitting a patient with twice the energy) - but you don't get a better resolution or a better SNR.
Eikka
not rated yet Apr 30, 2012
Well, it's not exactly the same thing. The problem is this:

http://en.wikiped..._pattern

What you'd want is individual photons hitting the target at random intervals in random locations, and preferably with random energies (full spectrum), so the interference patterns would smooth out. The energy of the speckle pattern would be spread out instead of peaking at certain frequencies that make it visible.

In other words, you want to approximate non-coherent light coming out of your laser, which produces coherent light.
antialias_physorg
not rated yet Apr 30, 2012
Exactly. But by adding noisy channels to one another you aren't reducing noise (i.e. you aren't increasing information garnered). You're just smearing the noise spectrum over a larger range.
Eikka
not rated yet May 01, 2012
Exactly. But by adding noisy channels to one another you aren't reducing noise (i.e. you aren't increasing information garnered). You're just smearing the noise spectrum over a larger range.


Which effectively increases your signal to noise ratio at the specific bands of interest that were previously obscured by the speckle, by moving the noise to other bands that you're not interested in.
Meyer
not rated yet May 01, 2012
Correct me if this is wrong. This was my interpretation... Imagine you emit and detect 100 photons. Due to the way the photons are generated, 90 of them are distributed evenly around the image, while an interference pattern causes 10 of them to be detected within a small area, overloading the sensor and causing a "speckle". If you generate the photons in a way that eliminates the interference, either those 10 photons can provide more detail to the rest of the image (as well as whatever was washed out by the speckle), or you can get an equivalent image (minus the speckle) using only 90 photons of energy.
antialias_physorg
not rated yet May 03, 2012
Correct me if this is wrong.

Consider yourself corrected.

Since you don't know which ten photons will cause the interference (and where) it's not really possible to either alter the emitter not to emit such photons or the receiver to have a sensor in that place that could handle a higher input (since you don't know where the place is going to be - The speckles keep moving, too).

The problem lies in the way lasers produce light. A laser is light amplification through STIMULATED emission of radiation. That measn that photons in the laser medium will cause oher photons to be emitted in a sort of avalanche effect. Occasionally that effect is high in one patricular region and you'll get a speckle.