Vibrational speckle tomography microscopy for fast intra-operative cancer tissue histopathology

Project Information

SpeckleCARS

Grant agreement ID: 101052911

DOI

10.3030/101052911

EC signature date 27 June 2022

Start date 1 January 2023

End date 31 December 2027

Funded under

• European Research Council (ERC)

Total cost

€ 2 726 936,00

EU contribution

€ 2 726 936,00

Coordinated by

CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE CNRS
France

Periodic Reporting for period 1 - SpeckleCARS (Vibrational speckle tomography microscopy for fast intra-operative cancer tissue histopathology)

Reporting period: 2023-01-01 to 2025-06-30

Summary of the context and overall objectives of the project

Histology is central to the diagnosis, staging, and treatment of cancers. It requires the removal of small regions of suspect tissues (biopsies) that are later sectioned and stained with haematoxylin and eosin (HE). However, histology suffers from major weaknesses: First the standard HE protocol cannot work in vivo and requires ex vivo biopsies. Second HE is labour intensive, time consuming and the final tissue sections inspection is possible only after 12 hours. A faster but less accurate protocol, known as extemporaneous histology, can be performed in 40 min and is used in an intra-operative context to guide surgeries.

The SpeckleCARS project aims to develop for the first time fast label free tomography microscopy with vibrational sensitivity to provide images with tridimensional improvement as compared to extemporaneous histology and in near-perfect concordance with conventional stained HE histology. This pioneering approach overcomes all the previous challenges: Using a wide field reflection scheme and label free contrast rules out all the drawbacks due to tissue removal and external labelling. It provides instantaneous intraoperative 3D histological data, improving the accessibility and accuracy of tumour detection, diagnosis and treatment. Combining the latest advances in 3D tomography reconstruction algorithms, high power laser systems together with key innovations in random speckle illumination and hyperspectral vibrational imaging, the SpeckleCARS project will provide for the first time volumetric histology imaging in real time over large (mm) fields of view without the need of tissue removal. To demonstrate the power of the developed technologies both ex vivo human and in vivo mouse cancer assessments will be conducted in collaboration with pathologists.
The SpeckleCARS approach is broadly applicable to all cancer types and to any tissue diagnostic requiring histological data, so the project breakthroughs will benefit many medical and biology applications

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The SpeckleCARS project has led to some research and technological achievements in the field of nonlinear microscopy using multiphoton processes to generate images.

Achievement 1:
Wide field CARS microscopy using speckle illuminations: The first research achievement has been to demonstrate for the first time CARS (coherent anti-Stokes Raman Spectroscopy) microscopy in wide field using speckle illuminations. This new scheme uses two speckle fields, the Pump and the Stokes, to generate the CARS process that is also a speckle field. By using a rapidly spatially varying Pump speckle and a fixed Stokes speckle we have shown that the phase of the CARS field can be randomized such that the CARS speckle behaves as an incoherent field. In this case it is possible to apply the scheme of Random Illumination Microscopy (RIM) that allows to access to optical sectioning and ultra-resolution. RIM captures different incoherent speckles, in this case CARS speckle, from the sample. Note that these various CARS speckles are created by varying the Stokes speckle while the pump one is still rapidly varying (as compared to the camera integration time). From the stack of the incoherent CARS speckle images, it is possible to obtain (i) a wide field image from the sample with (ii) an optical sectioning of few microns and (iii) a spatial resolution that is two times better than in a conventional wide field image. This ultra-resolution is reminiscent to structured illumination microscopy that illuminates the samples with a variety of grids to achieve a two-folds resolution improvement. In the case of speckle illumination, the structured illumination comes from the speckle itself. Finally, because we are using the CARS process the generated image has a chemical sensitivity and captures the chemical bond whose vibrational frequency equals the difference of frequency between the Stokes and pump beams.
This work has been published in Nature Photonics: https://www.nature.com/articles/s41566-023-01294-x
https://hal.science/hal-04217792v1/document

Achievement 2:
Chemical imaging of organoids: Stimulated Raman scattering is a coherent Raman process that is free of background and that can be used to generate images of tissues in agreement with classical histology. Using our knowledge in the virtual coloring of SRS images from the CH2 and CH3 chemical bonds we have been able to generate 3D images of organoids samples coming from the gastric system. Because SRS is a non-destructive imaging modality, we have been able to image and follow organoids over time during their growing process. We have shown that healthy and cancerous organoids develop in a different way. This work demonstrates the interest of coherent Raman imaging as a powerful imaging tool for organoid monitoring during growth and the possible exposition to drugs and treatments.
This work has been published in Nature Publishing Journal-Imaging: 10.1038/s44303-024-00019-1

Achievement 3:
Demonstration of wide field nonlinear microscopy with a high power optical parametric amplified laser system: Our high power laser source can be used to generate two-Photon Excited Fluorescence (2PEF), Sum Frequency Generation (SFG), and Hyperspectral Coherent Anti-Stokes Raman Scattering (H-CARS) microscopy. The source uses a high-power Ytterbium laser (100W) to pump two Optical Parametric Amplifiers (OPA) that can be rapidly tuned in frequency due to two acousto-optics narrow band filters that select the OPA seed beams from a white light. This unique source enables to perform rapid CARS spectroscopy (2 ms/spectrum) and nonlinear wide-field imaging, up to 3.3 frames/s for 2PEF and SFG and 0.3 hypercubes/s for H-CARS, over a field of view >300 × 300 µm2.
This work has been published in APL Photonics: https://doi.org/10.1063/5.0221283
https://pubs.aip.org/aip/app/article/9/9/096113/3312693/Dual-picosecond-fast-tunable-optical-parametric

Achievement 4:
Demonstration of a fast and agile acousto-optics delay line: In the SpeckleCARS project we are using CARS to perform chemical imaging. Interferometric CARS is a known scheme that allows to remove the nonresonant background in CARS. We developed a fast and agile acousto-optics delay line that can scan 50ps of delay in 20µs only. We demonstrated the advantages of this agile delay line in the context of impulsive stimulated Raman imaging in the low-frequency range (<200 cm−1). Besides fast imaging with a spectral resolution of 1.5 cm−1, we show that random-access delay may be exploited to selectively image molecular species at high speed
This work has been published in Optics Letters: https://doi.org/10.1364/OL.544222
under embargo until 08/07/2025

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The breakthrough and advances beyond teh state of the art is:

- Wide field CARS microscopy using speckle illuminations and the associated Nature Photonics publication (http://dx.doi.org/10.1038/s41566-023-01294-x) is at the heart of the speckleCARS project. This is a real breakthrough in the fields demonstrating that wide field CARS imaging is possible with significant advantages over conventional CARS imaging techniques.
It allows larger field of view, higher speed, less photodamage and enhanced spatial resolution CARS imaging. This was not possible before the ERC project.

SPeckleCARS general image