Clément GRAND soutiendra sa thèse intitulée « Compressive Raman Imaging, applications in environmental sciences » le mercredi 7 décembre 2022 à 10h00 dans l’amphi ROUARD, Campus St Jérôme à Marseille.
Cette thèse a été dirigée par Hervé Rigneault de l’équipe MOSAIC et le jury sera composé de :
Randy Bartels, Professeur, Colorado State University, USA (Rapporteur)
Eric Finot, Professeur, Université de Bourgogne, France (Rapporteur)
Julien Duboisset, Maître de conférence, Institut Fresnel, France (Examinateur)
Enora Prado, Chargé de recherche, IFREMER Bretagne, France (Examinatrice)
Andreas Zumbusch, Professeur, Universität Konstanz, Allemagne (Examinateur) - En visio
Sophie Brasselet, Directrice de recherche, Institut Fresnel, France (Présidente)
Hervé Rigneault, Directeur de recherche, Institut Fresnel, France (Directeur de thèse)
Résumé (anglais) :
Recent developments in Raman microspectroscopy have extended its application to biology, medicine and pharmaceutical sciences. A prime example is the significant improvements in imaging sensitivity and speed that have enabled to apply Raman to biomedical research. Raman imaging is nowadays a new imaging modality that can provide molecular level information in biological systems inaccessible by conventional optical techniques. Raman imaging provides label free, non-destructive, high chemical selectivity with superb optical resolution. However, Raman-based microspectroscopy still has its drawbacks. In general, the Raman signal is scattered and detected by a camera to ultimately form a hyperspectral image. This generates large volumes of data, as well as requires very long acquisition times. The high data volume inherent in Raman microspectroscopy is the major challenge that prevents dynamic spectral imaging fore live applications.
In this thesis, we use the Compressive Raman Technology (CRT), the development of which represents a significant advance in the field of Raman spectroscopy. CRT speeds up the measurement process and simultaneously simplifies the data analysis. CRT uses a programmable filter located in the spectral plane of a spectrometer. It is then possible to select a set of Raman lines specific to a chemical compound and detect them not with a camera, but with a faster single pixel detector. The technology is accompanied by a suite of algorithms that define the optimal filters to detect and identify known chemical species and recover their proportions. We apply first CRT imaging for the quantification of chemical species in the context of polymorph active molecular ingredient in pharmaceutical tablets. Second CRT imaging is applied for the detection of micro-plastics coming from natural environmental samples. In both cases we demonstrate the superiority of CRT imaging as compared to conventional Raman approaches.
LIEN ZOOM : https://univ-amu-fr.zoom.us/j/5196773417
ResearchGate
Flux RSS