Researchers Use Netflix Challenge Algorithm to Accelerate Biological Imaging

In OpticaThe multi-institutional research group of the Optical Society, specializing in high-impact research, says that a computer imaging approach called compressive imaging can increase the speed of imaging by reducing the amount of Raman spectral data acquired. They show imaging speeds of a few tens of seconds for an image that usually takes a few minutes to acquire and indicate that future implementations could reach speeds below one second.

The researchers accomplished this feat by acquiring only part of the data generally required for Raman spectroscopy, and then completing the missing information with an algorithm developed to search patterns in Netflix movie preferences. Although the algorithm did not win the $ 1 million Netflix award, it was used to meet other real needs, as in this case a better imagery biological.

"Although compressive Raman approaches have already been reported, they could not be used with biological tissues because of their chemical complexity," said Hilton de Aguiar, head of the research team at the University of California. Ecole Normale Supérieure in France. "We have combined compressive imaging with fast computer algorithms that provide the kind of images that clinicians use to diagnose patients, but quickly and without laborious manual post-processing."

Capturing biomedical processes

Raman spectroscopy is a non-invasive technique that does not require any sample preparation to determine the chemical composition of complex samples. Although the identification of cancer cells and the search for diseases by tissues are promising, image acquisition speeds that are too slow to capture the dynamics of biological samples are generally required. The processing of the mass of data generated by spectroscopic imaging also takes a lot of time, especially when analyzing a large area.

"With the methodology we have developed, we have simultaneously identified these two challenges: to increase speed and to introduce a simpler way to acquire useful information from spectroscopic images," said de Aguiar.

Optimize speed

To speed up the imaging process, the researchers made their Raman system more compatible with the algorithm. To do this, they have replaced the expensive and slow cameras used in conventional installations by a fast and economical digital micromirror device, called spatial light modulator. This device selects groups of wavelengths detected by a highly sensitive pixel detector, compressing the images as they are acquired.

"A very fast light spatial modulator allowed to acquire images and jump bits of data very quickly," said de Aguiar. "The spatial light modulator we used is much cheaper and faster than other options on the market, making the overall optical configuration economical and fast."

The researchers demonstrated their new methodology using a Raman microscope to obtain spectroscopic images from brain tissue and single cells, both of which have high chemical complexity. Their results showed that the method can acquire images at a speed of a few tens of seconds and achieve a high level of data compression, reducing them up to 64 times.

The researchers believe the new approach should work with most biological samples, but they plan to test it with more tissue types to demonstrate this experimentally. In addition to clinical tools, the method could be useful for biological applications such as the characterization of algae. They also want to improve the scanning speed of their system in order to achieve less than a second image acquisition.

Explore further:
Faster 3D Imaging Could Help Diagnose Cardiovascular and Gastrointestinal Diseases

More information:
F. Soldevila, J. Dong, E. Tajahuerce, Gigan S., B. De Aguiar, "Raman Rapid Compression Bio-imaging by Matrix Completion" Optica, 6, 3, 341-346 (2019). DOI: doi.org/10.1364/OPTICA.6.000341