Skip to main content
European Commission logo
Deutsch Deutsch
CORDIS - Forschungsergebnisse der EU
CORDIS
CORDIS Web 30th anniversary CORDIS Web 30th anniversary

Dual capillary waveguide endoscope

Periodic Reporting for period 1 - DARWIN (Dual capillary waveguide endoscope)

Berichtszeitraum: 2018-01-01 bis 2019-12-31

The overall goal of collaborative projects such as the BRAIN initiative and the Human Brain project is to understand how the dynamics of neural activity is transformed into the human cognition, emotion and behaviour. Optical techniques are unique to achieve this goal because of their scalability, allowing the study from sub-cellular compartments to the whole brain while providing high temporal accuracy, critical to resolve the millisecond-timescale neuronal activity signals. In the last decade, the development of proteins expressed in genetically modified virus enable cell-type specific imaging and manipulation of neural activity commonly referred as optogenetics. These studies have been carried out typically using an invasive approach where the skull is removed because the resolution and penetration depth of optical imaging in biological tissue is severely degraded by scattering, limiting the non-invasive approaches. For deep brain imaging with diffraction limited optical resolution, endoscopic approaches are required to avoid the loss of resolution due to scattering. But its typical cross section is between 0.6 and 1 mm, which can produce tissue damaged when inserted. One of the goals of the action is use capillary multimode waveguides that have a smaller cross section (~350um) to reduce the tissue damage.

However, coherent light propagating through a this small waveguides is seemingly randomized through fiber mode variations in phase velocity, leading to a granular speckle pattern at the far end of the fiber. Imaging through them therefore requires special methods. Another objective of the action is to explore new ways of calibration and imaging.

Because fluorescence imaging requires exogenous agents (dyes or genetically modified cells) to produce the contrast required to image, other complementary imaging techniques are being explored. One of this techniques is photoacoustic imaging that relies on endogenous contrast agents such as light absorption. Photoacoustic imaging an emerging multi-wave imaging modality that couples light excitation to acoustic detection via the photoacoustic effect (sound generation via light absorption), relies on detecting ultrasound waves that are very weakly scattered in biological tissue. The other goal of the action is to not only design an ultra-thin endoscope for fluorescence imaging, but also for photoacoustic imaging and take advantage of the hollow core of the capillary to place a device capable of detecting ultrasound waves providing images in both modalities. This combination can enable simultaneous monitoring of fast spatio-temporal neurodynamics and vasular hemodynamics, hence providing unprecedented capabilities for neural activity detection.

The project demonstrated the feasibility of building a small endoscope capable of image using complementary imaging techniques and perform proof-of-concept experiments.
"The fellow has designed, built and optimized an experimental setup that is capable of imaging through capillary waveguides and multimode fibers using two different imaging modalities: fluorescence and photoacoustic imaging. Besides developing the experimental setup, the fellow had perform simulations for new imaging techniques using the granular speckle patterns produced at the output of the capilllary and the mulitmode fiber. After the simulations showed promising results, their experimental implementation with the setup matched the predicted viability of the proposed method. This novel imaging approach reduce the complexity of the experimental setup and reduce the number of measurements required to form an image at expenses of increasing the computational complexity in the image reconstruction.

Three types of proof-of-principle experiments were performed in this work, namely photoacoustic microscopy alone, fluorescence microscopy alone, and hybrid photoacoustic/fluorescence microscopy. Two types of samples were used to illustrate the photoacoustic imaging capability of the setup. As a well-controlled test sample, we used an absorbing micro-structure photoplotted on a polymer film shown on Fig. 1 a (""power-on"" logo). To further demonstrate the performance of the system on a more relevant biological sample, we used the same system to obtain photoacoustic images of red blood cell shown in Fig. 1c. Fig. 1b and Fig. 1d shown the reconstructed objects using the photoacoustic signals of each speckle pattern projected.

Two types of fluorescent samples were imaged with a similar experimental setup. Fig. 2 a shows a reference widefield fluorescent image of 4 µm orange beads. The image from fluorescence collected at the input/proximal side of fiber is shown in Fig. 2 b, where the complex distribution of beads is well-recovered while preserving the boundaries of both individual and clustered beads. We also performed imaging of red fluorescent retrobeads (0.05 - 0.2µm) from Lumafluor, microinjected into the dorsomedial striatum (DMS) of a mouse brain, which was then sliced and mounted on a microscope slide. As for the first sample, Fig. 2 c shows a reference widefield fluorescent image of the sample. Fig. 2 d shows the corresponding image reconstructed with our approach, and also clearly demonstrates the recovery of individual clusters of retrobeads in neurons.

Finally, we performed an experiment where we obtain both the photoacoustic and fluorescence images of the same sample. The same set of speckle patterns from a unique calibration procedure was used for both photoacoustic and fluorescent imaging. We used a diluted solution composed of red blood cells and 11μm diameter fluorescent particles (Nile red) in PBS. Figure 3a shows the widefield microscope image of the sample using incoherent illumination through the MMF where two fluorescence particles and one red blood cell can be identified. Figure 3b shows the false-color reconstructed photoacoustic image (red) and fluorescence image (green) from the set of signals corresponding to 2048 speckle patterns. The photoacoustic image clearly shows the ability to resolve single red blood cells and it does not have any spurious signal coming from the fluorescence particles as expected. In the fluorescence case, the signal recorded by the PMT only has contributions from the fluorescence particles.

The fellow has attended and presented the work in more than 10 international conferences, has given several invited seminars in 3 different universities and has participate in an outreach activity in a pub in Grenoble."
During the action, the fellow has developed an ultra-thin system capable of performing both fluorescence and photoacoustic microscopy based on an optical multimode fiber (MMF) and an optical fiber hydrophone and capable of combined photoacoustic and fluorescence microscopy. A reference-free calibration, using a set of pre-recorded speckle intensity patterns combined with a reconstruction algorithm, enables compressive speckle imaging with optical resolution determined by the fiber numerical aperture.

After the action is finished, the fellow will work as the CEO of a startup company found in United States, in collaboration with other 4 professors. This company will develop a prototype of an ultrathin endoscope based on multimode fiber for imaging the brain activity. The experience acquired by the fellow during the action in photoacoustics and fluorescence have been critical to have this unique opportunity.
Experimental results of hybrid (fluorescence and photoacoustic) microendoscopy
Experimental results for photoacoustic microendoscopy
Experimental results of fluorescence microendoscopy