#	Pagina
attuale pagina	/open-h2020/projects/194544/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 3 - VIBRA (Very fast Imaging by Broadband coherent RAman)

Teaser

Summary

Raman microscopy provides functional information on tissues and cells. It is a non-contact, non-invasive and non-destructive method, as it does not require any sample labelling/staining. Every component of a biological specimen is characterized by a vibrational spectrum that reflects its molecular structure and provides an endogenous and chemically specific signature that can be exploited for its identification. The main drawback of spontaneous Raman is its very weak scattering cross section, 10-12 orders of magnitude lower than absorption. This makes it difficult to: (i) separate the weak Raman-shifted light from the intense elastic scattering and from sample and substrate fluorescence; (ii) probe dilute species; (iii) image dynamical processes in living organisms, due to the long integration times needed (up to several minutes for a single spectrum in one sample position).
Coherent Raman Scattering (CRS) can overcome these hurdles, by setting up a coherent superposition of vibrational responses. When the frequency difference between two narrowband picosecond pulses (pump and Stokes) matches a characteristic vibrational frequency, all the molecules are resonantly excited. This enhances the Raman signal by many orders of magnitude, enabling much higher imaging speeds. CRS microscopy provides further advantages: (i) being a nonlinear microscopy technique, the signal is generated only in the focal volume, thus allowing three-dimensional imaging; (ii) working out of resonance, it minimizes photo-damage to biological samples; (iii) compared to fluorescence microscopy, CRS is label-free and does not require sample labelling with exogenous fluorescent markers that often perturb the biological functions.
State-of-the-art CRS microscopy has reached extremely high imaging speed (up to the video rate), but it provides limited chemical selectivity. Most of the currently available CRS systems, in fact, deliver information at just a single vibrational frequency. This is not sufficient to distinguish the different components within complex heterogeneous systems, such as cells and tissues, with spectrally overlapped chemical species. Another unsolved issue in current state-of-the-art CRS microscopy is the narrow spectral coverage, typically restricted to the high wavenumber window (2700-3600 cm-1) displaying a high density of vibrational oscillators. Here the Raman spectrum is a superposition of broad features arising from the stretch of hydrogen bonds in long-chain hydrocarbons, lipids and water molecules. Few reports exist on probing the more informative fingerprint region (600-1800 cm-1), due to the weaker Raman signatures.
The ground-breaking goal of VIBRA project is to develop broadband high-speed CRS techniques, combining the highest molecular information content of spontaneous Raman with the high imaging speed of Coherent Raman, acquiring functional images at a speed as close as possible to the video-rate. This will have a revolutionary impact in biology and medicine: it will allow researchers and doctors without a specific knowledge in lasers and optics to routinely visualize functional properties of cells and tissues in vivo that have never been observed before with such a combination of detailed bio-chemical specificity and real-time non-invasive imaging.

Work performed

(1) We have developed a complete spectroscopic system and microscope for performing nonlinear imaging in a multimodal approach, combining two-photon excitation fluorescence (TPEF), coherent anti-Stokes Raman scattering (CARS) and stimulated Raman scattering (SRS). This instrument can bring a very rich biological information on the sample under study. Up to now, NLO microscopes were based either on two synchronized Ti:sapphire oscillators or a bulk solid-state oscillator driving an optical parametric oscillator. The complexity and cost of the overall apparatus, from the laser source to the detection systems, has so far hindered its mainstream use in biology. We have greatly simplified the excitation scheme and reduced the costs and maintenance of the system the experimental setup, employing a fiber laser source. A pair of galvanometric mirrors scans the beam over the sample area at high speed. SRS/CARS spectra can be rapidly recorded in the 2800-3100 cm-1 Raman region, thus providing a detailed chemical information on the sample.
(2) We demonstrated a new approach to broadband SRS, based on time-domain Fourier transform (FT) spectroscopy. While broadband CARS has been reported, recording broadband SRS spectra is very challenging: so far, multiplex SRS has only been performed with narrowband pulses, by rapidly scanning the pump-Stokes frequency detuning. Our approach blends the very high sensitivity of single-channel lock-in balanced detection with the spectral resolution afforded by FT spectroscopy. A narrowband pump pulse and a synchronous broadband Stokes pulse illuminate the sample. The transmitted Stokes is sent to a homemade Fourier transform spectrometer based on a passive birefringent delay line, allowing us to create two replicas of the Stokes pulse, with perpendicular polarizations, whose delay can be finely controlled with exceptional path-length stability and reproducibility. Our spectral resolution is of the order of 30 cm-1, limited by the scan range of the interferometer, but could be easily improved, if required by the experiment, using a different design, till the 15 cm-1 limit set by the pump pulse bandwidth, at the expense of an increased acquisition time. We applied the technique to SRS microscopy. We imaged a mixture of polymethyl methacrylate (PMMA) and polystyrene (PS) spherical beads with 6-Î¼m and 3-Î¼m diameter, respectively. The measured three-dimensional dataset as a function of sample position and Raman shift has been analyzed using a Multivariate Curve Resolution-Alternating Least Square algorithm. It decomposes the experimental dataset in a set of spectra and concentration vectors. In this way, we correctly identified two principal components only, with spectra perfectly matching those of PMMA and PS. Our novel approach paves the way to high speed chemical identification of biomolecules via their broadband coherent Raman response.
(3) We demonstrated a novel approach to broadband (multiplex) SRS, based on Photonic time stretch (PTS) spectroscopy. PTS is a powerful technique for single-channel measurement of optical spectra at high repetition rates. It consists in temporally stretching the pulse, typically by a long optical fiber, to a duration of a few nanoseconds, so that it can be accurately sampled by a high-frequency analog to digital converter. By calibrating the dispersion introduced by the optical fiber, each point of the sampled temporal profile can be uniquely associated with a wavelength. Spectra can then be measured at repetition rates up to tens of MHz. Using a laser system running at 80-kHz repetition rate, we demonstrated the acquisition of broadband SRS spectra across the CH stretching band (2700-3250 cm-1) with sensitivity well below 10-4. By a proper optimization of the excitation source, we anticipate a significant increase of the detection bandwidth and the acquisition speed.

Final results

We are pushing the technology of coherent Raman microscopy toward higher speed and larger bandwidths. The aforementioned results listed above represent a great progress beyond the state of the art. The impact in the society at the end of the project is the demonstration of the great potential of this instrument for cell and tissue identification, via rapid microscopy imaging.
During brain tumour surgery, leaving behind cancerous tissue can allow it to spread, while removing healthy tissue can cause neurologic deficits. It is thus essential to find the surgical margin, which is not visible by eye. Furthermore brain matter is a very soft tissue: during surgery it moves and changes shape so that it is often hard to refer to previously collected images (via magnetic resonance or computed tomography scans). My vision beyond the VIBRA project is that the developed vibrational microscopes could be used in the future (in 5-10 years) to perform â€œvirtual histopathologyâ€: rapid in-vivo intraoperative assessment of brain tissue, safely and accurately identifying tumours and tumour boundaries.