Skip to Main content Skip to Navigation

Inversion de modèle et séparation de signaux de spectroscopie optique pour la caractérisation in vivo de tissus cutanés

Abstract : The detection of pre-cancerous biological tissues during clinical diagnosis is made possible by the development of optical spectroscopic methods, also known as optical biopsy. These methods make it possible to identify changes in the metabolism and structure of biological tissues. In this thesis, optical measurements are performed on skin tissues using an optical device that integrates two modalities: spatially resolved diffuse reflectance and autofluorescence. Spectra exploitation involves solving a non-linear inverse problem to estimate the optical properties using a light-tissue interaction model. From these optical properties, it is possible to characterize the condition of the tissue, healthy or abnormal. The first contribution of the thesis is to propose an inversion method that integrates a fast Monte Carlo model and a cost function managing two spectral modalities in order to improve the accuracy of the estimation of optical properties, regardless of the geometry of the probe. The second contribution is to study the kinetics of fluorophores present in tissues by a source separation approach, which does not require the use of a Monte Carlo model. In this approach, a collection of spectra acquired over time is analyzed together, with few a priori assumptions about the shape of the source signals. The sensitivity of this approach is validated by the use of optical clearing that increase tissue transparency by decreasing photon diffusion and absorption in tissues, thus providing access to more accurate in-depth information.
Complete list of metadata

Cited literature [166 references]  Display  Hide  Download
Contributor : Thèses UL Connect in order to contact the contributor
Submitted on : Monday, June 22, 2020 - 12:15:08 PM
Last modification on : Wednesday, November 3, 2021 - 4:47:15 AM


Files produced by the author(s)


  • HAL Id : tel-02877382, version 1


Prisca Rakotomanga. Inversion de modèle et séparation de signaux de spectroscopie optique pour la caractérisation in vivo de tissus cutanés. Automatique / Robotique. Université de Lorraine, 2019. Français. ⟨NNT : 2019LORR0329⟩. ⟨tel-02877382⟩



Record views


Files downloads