Optics.org
Optics.org
daily coverage of the optics & photonics industry and the markets that it serves
News
Menu
Research & Development

Fiber-optic probe promises endoscopic cancer diagnosis

03 May 2017

Compact device from Jena team offers alternative to conventional biopsies.

Accurate diagnosis of cancer typically requires removal of a tissue sample for analysis, but optical alternatives to these biopsy procedures would be faster, less invasive and potentially more precise.

A hand-held fiber-optic probe developed by a Jena-based project team could offer a promising route towards endoscopic cancer diagnosis, and remove the need for external tissue staining.

The multi-modal device, described in a paper published in Optica, uses an ultrafast laser to create nonlinear optical effects in tissue, effects that can reveal the presence of cancer and other diseases.

"We hope that, one day, multi-modal endoscopic imaging techniques could help doctors make quick decisions during surgery, without the need for taking biopsies, using staining treatments or performing complex histopathological procedures," said Jürgen Popp of the Leibniz Institute of Photonic Technology (IPHT), one of the project partners.

"The new probe serves as a miniaturized microscope that uses near-infrared lasers to investigate tissues. Different components of biological tissue react differently to the excitation lasers, and their unique response gives us information about the molecular composition and morphology within the tissue."

The diagnostic value of nonlinear optical imaging modalities such as coherent anti-Stokes Raman scattering (CARS), second harmonic generation (SHG), and two-photon excited auto-fluorescence (TPEF) microscopy has been known for some time, with flexible fiber probes offering a promising way to carry out endoscopic examinations in difficult to reach areas.

As the team's paper notes, implementation of SHG and TPEF is relatively straightforward. However, the generation of a background-free CARS signal in situ has proven more challenging, due to the high pulse intensities required and the need for the coordination of two laser pulses of different wavelength.

Ten thousand elements

The group at IPHT and Friedrich Schiller University of Jena has now tackled this hurdle by designing a multi-core imaging fiber consisting of ten thousand coherent light guiding elements, preserving the spatial relationship between the entrance and the output of the imaging fiber. The diameter of the individual cores ranges from 2.2 to 3.7 microns.

An important factor in the probe's reduced size is the use of gradient index, or GRIN, lenses to focus the laser light. GRIN lenses can be made very small because they focus light through continuous refractive index changes within the lens material, and project partners Grintech supplied GRIN lenses only 1.8 mm in diameter for the Jena probe.

When tested on human skin tissue, the probe was able to acquire separate CARS, SHG and TPEF images with a resolution of 2048 by 2048 pixels in a scanned area of 300 by 300 microns. This is sufficient for identifying tumor borders, and the probe can be moved over the tissue surface to get an overview of the affected area.

In practice, moving one end of the probe across a sample and transferring the acquired images to the other end is "not a trivial task, since the cores of the imaging fiber differ in size and shape, hindering efficient and homogeneous coupling of the excitation lasers," said Popp. "Furthermore, we had to deal with unwanted effects like different wavelengths interacting inside the fiber and core-to-core light coupling."

The next steps for the project include algorithms to improve the quality of the multi-modal images, which at present appear pixelated because of the structure of the multi-core imaging fiber. Tests in animal models and with patients in a clinical setting should follow.

Photon Engineering, LLCAUREA TECHNOLOGYSACHER LASERTECHNIK GMBHEKSPLA uabCobolt ABOmicron-Laserage Laserprodukte GmbHInfraTec GmbH
Copyright © 2019 SPIE EuropeDesigned by Kestrel Web Services