Thèse
Publié le 2 mars 2018 | Mis à jour le 19 septembre 2018

Laure ALSTON 2014-2017 - PhD director: David Rousseau, co-directors: Bruno Montcel and Matthieu Herbert

équipes CREATIS-5, LabHC

Title: Spectroscopie de fluorescence et imagerie optique pour l'assistance à la résection de gliomes : conception et caractérisation de systèmes de mesure et modèles de traitement des données associées, sur fantômes et au bloc opératoire

Gliomas are infiltrative tumors of the central nervous system that account for more than 50% of malignant primitives brain tumors. The WHO classify them into 4 groups [1] and glioblastomas are considered as the strongest ones, highly infiltrative, with a survival rate not much higher than 14 months [2]. Studies showed that the survival rate was linked with the extent of resection [3, 4, 5] but a tradeoff between the resection of a maximum amount of tumor cells and the preservation of functional areas needs to be found during neurosurgery. Thus, interventional imaging is clearly relevant to help neurosurgeon intraoperatively.
The fluorescence of protoporphyrin IX (PpIX) is widely used to guide tumor resection through a surgical microscope since PpIX accumulates in tumor cells and emits reddish fluorescence under violet/blue excitation [6] (see fig. 1) but the sensitivity of this technic is limited when it comes to tumor margins. If fluorescence spectroscopy of PpIX is currently studied [7, 8], those studies only consider PpIX with a spectrum peaking at 634 nm. We demonstrated in vitro and ex vivo the presence of a second peak around 620 nm, attributed to a second state of PpIX and we showed that the use of the two states can help discriminate strong high-grade gliomas towards their isolated cells and towards low grade gliomas[9]. We now want to apply this in vivo to get new parameters to discriminate the different kinds of gliomas among themselves and against healthy tissue.
In parallel, to identify functional areas, one can either study the electrical response of neuronal cells or the haemodynamic response of the brain [10]. Indeed, we assume that brain activity leads to haemodynamic variations which latter yield small color variations however visible in processed color images, thus allowing identification.
For now, the two aspects are developed in parallel, with two different protocols that both need to be performed in the operating room.
  • Auteur(s)
    Laure Alston