High Optical Resolution for Unlabelled Samples – HORUS

The project relies on the use of advanced reconstruction methods, from the inverse problems domain, to provide high-resolution, low-noise, index-calibrated images, as well as to accelerate data acquisitions to enable the use of TDM to observe quickly-moving, living biological specimens. To this purpose, the project associates 3 teams, specialists in their respective fields. The MIPS-Mulhouse laboratory has been developing TDM technology for about ten years, and has already achieved several world firsts in this field. This group will bring its experience in instrumentation, unique in France in this domain, with its own prototype, but also with the semi-rapid system that it has already developed for the IGBMC-Strasbourg, for the study of viral infections by HIV, coupled with confocal fluorescence microscopy, and scanning electron microscopy. LaHC-St Etienne works on inverse problem solving and co-design of imaging devices, with a strong experience in unconventional imaging devices. In particular, the group has developed rigorous statistical methods based on inverse problems for digital holography. As TDM is in fact an extension of digital holographic microscopy, the know-how of this group will allow for a deep rethinking of the image reconstruction approaches up-to-now used in the domain. The IGBMC Lamour-Ruff team focuses on the molecular mechanisms governing the transport, processing and modification of nucleic acids. Its current targets are nucleoprotein complexes involved in the integration of retroviral DNA and eukaryotic DNA topoisomerases. This team will use the system developed by MIPS for its own research, and will also be supported by the Imaging Center at IGBMC, an IBiSA platform that provides IGBMC researchers with extensive equipment.

A rapid microscope was developed at IRIMAS using optical prototyping elements.
This instrument has been cloned for use on an imaging platform at the IGBMC.
Two new image reconstruction approaches have been developed, based on a fast homotopy, and on the GSURE criterion via the LaHC-IRIMAS collaboration.
Two studies on the optimization of the TDM illumination have been performed and published

Adapting the new reconstruction methods for easier handling by end users.
Validating the system at the IGBMC for HIV studies
Finalizing the addition of tomographic imaging adapted to birefringent specimens

Denneulin L. , Momey F., Brault D., Debailleul, M., Taddese A. M., Verrier N., Haeberlé O. GSURE criterion for unsupervised regularized reconstruction in Tomographic Diffractive Microscopy. J. Opt. Soc. Am. A 39, pp. A52-A61 (2022)

J.-B. Courbot And B. Colicchio, A Fast Homotopy Algorithm For Gridless Sparse Recovery Inverse Problems, Vol. 37, p. 025002 (2021)

Asemare M. Taddese, N. Verrier, M. Debailleul, J.-B. Courbot, and O. Haeberlé, “Optimizing Sample Illumination Scanning in Transmission Tomographic Diffractive Microscopy”, Appl. Opt. 60, pp. 1694-1704 (2021)

Asemare M. Taddese, N. Verrier, M. Debailleul, J.-B. Courbot, and O. Haeberlé, “Optimizing Sample Illumination Scanning in 4Pi and Mirror-Assisted Tomographic Diffractive Microscopy” Appl. Opt. 60, pp. 7745-7753 (2021)

Three-dimensional optical microscopy is an invaluable tool in many fields, particularly in biology, thanks to its unique properties for imaging living specimens. In recent years, tremendous progress has been made with the development of structured illumination techniques, PALM / STORM, and STED. However, when fluorescence labeling cannot be performed, or is to be avoided, the resolution of optical microscopy is still limited to about 200 nm in practice.

Tomographic Diffractive Microscopy (TDM) is a recent technique that computes the image of the specimen from the electromagnetic field it diffracts. It allows for observing unmarked specimens in 3D, with twice better resolution than with conventional optical microscopy, and gives access to a new information: the 3D complex optical index distribution within the observed sample.
This technique begins spreading, but, at present time, still suffers from certain limitations. The most commonly used reconstruction assumptions (Born approximation, for example) may not be satisfied. Digital reconstructions are constrained. This limits their flexibility, and makes the quantitative measurements very sensitive to the conditions of observation (noise ...). Finally, a large number of acquisitions is needed to reconstruct high-resolution, high-quality images.

The HORUS project aims at developing practical solutions to overcome these problems, and to facilitate the adoption of TDM, especially for use in biological investigations. The project relies on the use of advanced reconstruction methods, from the inverse problems domain, to provide high-resolution, low-noise, index-calibrated images, as well as to accelerate data acquisitions to enable the use of TDM to observe quickly-moving, living biological specimens.

To this purpose, the project associates 3 teams, specialists in their respective fields.
The MIPS-Mulhouse laboratory has been developing TDM technology for about ten years, and has already achieved several world firsts in this field. This group will bring its experience in instrumentation, unique in France in this domain, with its own prototype, but also with the semi-rapid system that it has already developed for the IGBMC-Strasbourg, for the study of viral infections by HIV, coupled with confocal fluorescence microscopy, and scanning electron microscopy.
LaHC-St Etienne works on inverse problem solving and co-design of imaging devices, with a strong experience in unconventional imaging devices. In particular, the group has developed rigorous statistical methods based on inverse problems for digital holography. As TDM is in fact an extension of digital holographic microscopy, the know-how of this group will allow for a deep rethinking of the image reconstruction approaches up-to-now used in the domain.
The IGBMC Lamour-Ruff team focuses on the molecular mechanisms governing the transport, processing and modification of nucleic acids. Its current targets are nucleoprotein complexes involved in the integration of retroviral DNA and eukaryotic DNA topoisomerases. This team will use the system developed by MIPS for its own research, and will also be supported by the Imaging Center at IGBMC, an IBiSA platform that provides IGBMC researchers with extensive equipment.

The project therefore associates three very complementary teams, to bring significant progress in instrumentation (optimization and acceleration of acquisitions, self-calibration of data), image reconstruction / processing (denoising, reconstruction by inverse problem approach), applied to a practical biological case of great interest (HIV infection).
Associated with a recognized imaging platform, we believe that this project could have a strong scientific impact on all areas using optical microscopy for the study of transparent samples (biology, petrography, crystallography...), but also, in a longer term, in technological areas such as micro-optical inspection, or for quality control.