Finding treasures with physics: the fingerprint matrix

Left: Artistic impression of metal spheres buried in small glass beads. Middle: Conventional ultrasound picture. Right: With the new technology, the positions of the metal spheres can be precisely determined.
Image Credit: © TU Wien / Arthur Le Ber

How do you find objects buried in sand or hidden in thick fog? A team from the Institut Langevin (Paris) and TU Wien (Vienna) has developed an astonishing method.

Can we reveal objects that are hidden in environments completely opaque to the human eye? With conventional imaging techniques, the answer is no: a dense cloud or layer of material blocks light so completely that a simple photograph contains no information about what lies behind it.

However, a research collaboration between the Institut Langevin and TU Wien has now shown that, with the help of innovative mathematical tricks, objects can be detected even in such cases – using what is known as the ‘fingerprint matrix’. The team tested the newly developed method on metal objects buried in sand and in applications in the field of medical imaging. A joint publication on this topic has just appeared in the journal Nature Physics.

 A water tank with glass beads under which a metal sphere is hidden
Image Credit: © CNRS / TU Wien

Seeing and hearing means wave scattering

Whether we take a normal photo or use ultrasound to look inside the body – from a physical point of view, the same thing always happens when we create an image: a wave is sent to an object, the object reflects part of the wave, and the reflected portion reaches our eye – or a measuring device. This reflected wave can be used to determine where the object is located.

However, this normally only works if the object's surroundings are sufficiently transparent. “Otherwise, for example in a dense cloud or in murky water, the phenomenon of multiple scattering occurs,” explains Prof. Stefan Rotter from the Institute of Theoretical Physics at TU Wien. The wave is scattered not only by the object to be imaged, but also by the surrounding environment – often many times over, so that only a greatly altered wave can be registered, in which the object being sought can no longer be recognized.

“Instead of the object, all you see is a diffuse fog – this is a fundamental problem of imaging techniques, from sonar in submarines to imaging techniques in medicine,” says Lukas Rachbauer, one of the co-authors of the study.

First the fingerprint, then the image

To overcome this problem, the French-Austrian research team developed a novel method: first, a specific object is examined in an interference-free environment. Each object scatters waves in a very specific, characteristic way. This wave scattering fingerprint of the object can be described mathematically by a matrix – the so-called scattering matrix.

The object is then hidden in a highly scattering medium – for example, buried in sand. “When ultrasonic waves are sent into this sand, they are scattered by the sand, but some of the sound penetrates so far into the sand that it is also scattered by the buried object,” says Stefan Rotter. “We cannot see the object, but the backscattered ultrasonic wave that hits the microphones of the measuring device still carries information about the fact that it has come into contact with the object we are looking for in the sand.”

If, on the one hand, you know the unaltered scattering matrix, the ‘fingerprint matrix’ of the object, and, on the other hand, you measure the wave scattering generated by the hidden object in a multiple-scattering medium, then you can calculate the position of the object using a mathematical method developed by the research team.

“From the correlations between the measured reflected wave and the unaltered fingerprint matrix, it is possible to deduce where the object is most likely to be located, even if the object is buried,” explains Stefan Rotter.

Steel balls in sand and medical markers

The method was tested with steel balls in sand, but also in medical applications: to monitor the recurrence of breast cancer, so-called lesion markers are used, which are often difficult to image because they are overlaid by scattered signals. With the new method, they were easy to locate. 

In addition, fingerprint matrix technology was used to measure muscle fibers – which is particularly important for the diagnosis of heart and muscle diseases.

“The concept of the fingerprint matrix is very generally applicable – not only for ultrasound, but also for detection with light,” says Stefan Rotter. “It opens up important new possibilities in all areas of science where a reflection matrix can be measured.” Some objects also change their ‘scattering fingerprint’ when certain physical parameters change – such as pressure or temperature. Such variables could be measured from a distance using the new method. This would be particularly exciting, for example, in the measurement of the human brain, where waves have to penetrate the highly scattering skull.

Additional information: In collaboration with CNRS Innovation and the Patent & License Management Department at TU Wien, the fingerprint method has now also been submitted for patenting and taken up by a medical technology company.

Published in journal: Nature Physics

Title: Detection and characterization of targets in complex media using fingerprint matrices

Authors: Arthur Le Ber, Antton Goïcoechea, Lukas M. Rachbauer, William Lambert, Xiaoping Jia, Mathias Fink, Arnaud Tourin, Stefan Rotter, and Alexandre Aubry

Source/Credit: Technische Universität Wien

Reference Number: phy100525_01

Privacy Policy | Terms of Service | Contact Us