Image the Cortex

[Enthalpy]

Researchers at the Amsterdam University use time-resolved Raman spectroscopy to image objects embedded in non-transparent materials. They target explosives hidden in clothes or luggage (an obsession these days).

 

Time-resolution (read: short time) allows to discriminate at the detector between light diffused by the non-transparent material and by the embedded target, whose signal is fainter but strikes the detector from a different distance hence at a different time. Raman effect, as it scatters light back at a different wavelength not produced by the surrounding medium, allows further discrimination.

 

Under "non-transparent" I expect things like cloth or leather or plastic, not aluminium sheet.

 

A brief description in Spectroscopy Europe, vol 24 No 1 of February-March 2012.

 

I wish this nice technique be developed for purposes I feel more useful, especially imaging the brain's cortex within the skull. This is done by MRI or X-ray tomography up to now, which needs huge and costly apparatus to give not very clear images. Pictures of broken bones, clearer than by X-Rays, would be useful as well.

 

Beyond diffusion, the surround medium (the skull) also deforms the picture by refraction, but corrective software already exists, especially if several pictures are taken from different angles.

 

=================================================

 

A different way to discriminate light from the target and from the surrounding medium would have directional sensors AND illumination source. By putting them apart, light scattered from different depths arrives at the sensor from different directions, which for instance a CMOS focal plane can discriminate.

 

 

High contrast imaging is known from astronomers for instance, as they image planets around remote stars.

The directional source is swept to acquire a complete image. If the detector is a 2D plane, the source beam can be flat and scan only one angle, saving time. If the beam is narrow and scans two angles, the detector can be a line.

 

Again, pictures from different angles of view can allow software to compensate for refraction by the surrounding medium.

 

Raman scattering would further help discriminate the target from the surrounding medium and offer useful information about the materials, and make constrast products more efficient.

 

Marc Schaefer, aka Enthalpy

 

===================================

 

The method with directional sensor and source might be the one used by secret services for their mind-reading machines, in combination with software to compensate for head movements and some automatic learning software that identifies each area of the cortex.

 

Though, as such machines existed 30 years ago, I suppose they used an imaging radar, since this was available technology then. The necessary angular resolution looks feasible from few metres.

Time-resolved Raman is even less plausible in the late 70's.

 

Since the worst possible people have already this technology, I ask researchers to develop it for useful purposes as well, like medicine.

Four layers of space blanket, which are opaque from LW to visible, protect against such machines anyway.

[Xittenn]

There is plenty of research into the imaging of neural transmitters and their 'biological emissions' using various forms of Raman Spectroscopy, but most of these are invasive in one way or another. I don't see why you are going on about the secret services and brain imaging scanners as this would be no more effective than taking a picture of the skull cap itself under the current restrictions of medical technologies. Your post is really off on some weird tangent, I think that if you did some proper research into the matter you would find what it is you are hoping to see--quite in contrast to what your last three paragraphs suggest.

Edited March 22, 2012 by Xittenn

[Enthalpy]

Non-invasive methods work right now to map the cortex' activity: MRI (magnetic resonance imaging), PET (positron emission tomography), MEG (magneto-encephalogram). MRI and PET map the flow of blood in the cortex (not the less abundant neurotransmitters) which varies locally according to the activity.

 

The time-resolved Raman build by the researchers in Amsterdam does show hidden objects, like explosives in a bottle. I only suggest to show the cortex within the skin and skull, especially the local blood flow. This is not a picture of the skull.

 

Scattering of light is a very serious limit to pictures made through the flesh if using visible or near-infrared light. The explosive detection suggested by the research paper is far less hindered by scattering than a picture.

 

I had falsely hoped Raman would completely suppress scattering, as the Raman frequency change happens only once - but unfortunately, light can be scattered in the flesh both before or after the Raman effect.

 

This must be a big advantage of time-resolved Raman over the process I suggest, with illumination and observation from different angles.

• In my proposal, light scattering followed or preceded by Raman appear like a Raman hapening at a different depth without scattering.
• At time-resolved Raman, it takes two successive scatterings compensating an other so the detected light comes from the proper direction: less probable, especially if the in- and outcoming rays are offset by more than one cell diameter.

Scattering is so intense in flesh for red light that it should hamper even time-resolved Raman, but longer wavelengths should improve that - especially if longer than a cell diameter, which would be a somewhat longer wave than thermal infrared. Less comfortable: the frequency would be similar to a typical Raman shift.

 

In conditions where scattering isn't extremely strong, an different or complementary approach would let the light source and the imager move with respect to the observed object and add the light coming from the same voxels - a bit like we move our head to observe an object through a dirty window. Over many frames, non-scattered light adds up more efficiently to result in a clearer picture.

 

Marc Schaefer, aka Enthalpy

Edited May 11, 2012 by Enthalpy