Computational Raman microspectroscopy : faster and more sensitive

Vous êtes cordialement invités à la conférence "Computational Raman microspectroscopy : faster and more sensitive" de Hilton B. de Aguiar, Department of Physics, ENS, Paris, France, organisée par l'Institut Charles Sadron.

Résumé :

Raman imaging is recognized as a powerful label-free microscopy technique, providing high spatial resolution with superb molecular selectivity, exploiting the intrinsic vibrational spectra of molecules as a fingerprint. However, high speed and high sensitivity Raman imaging is often very challenging to achieve for various reasons. On the one hand, the weak Raman scattering effect requires sensitive camera, which nevertheless have additive noise at high speeds. On the other hand, high chemical selectivity comes at a price of acquiring huge spectroscopic imaging data sets. In particular, the hyperspectrum of Raman-based microspectroscopies preclude high bandwidth (e.g. video-rate) or large-scale imaging, aspects that are extremely important for the promising applications of Raman microscopy in biology and biomedicine.
In this presentation, I will introduce an emerging modality in Raman imaging: computational Raman microspectroscopy. As the name suggests, computational Raman exploit efficient algorithms by combining established concepts of chemometrics, novel signal processing schemes, and, to a certain extent, novel experimental sampling schemes.
I will show two examples exploiting the concept of computational Raman imaging: (i) using chemometrics analysis to obtain high sensitivity chemical images down to a single lipid monolayer [1], and (ii) introduce the recent compressive Raman framework [2] enabling the fastest chemical imaging in biomedical applications (few seconds imaging speeds). In particular, I will emphasize how aspect (i) allows for a thermodynamics analysis at the sub-micron scale.
References
[1] Donaldson Jr. and de Aguiar, J. Phys. Chem. Lett. 9 (2018), pp 1528–1533

Les personnes souhaitant rencontrer Hilton B. de Aguiar sont priées de prendre contact avec Carlos Marques (tel. 03 88 41 40 45 ou mail carlos.marques@ics-cnrs.unistra.fr)