PostDoc Project
Title:
Combination of diffuse reflectance spectroscopy and autofluorescence spectroscopy with line-field confocal optical coherence tomography for skin cancer in vivo diagnosis: spectroscopic data processing and analysis, light-tissue interaction modelling and experimental validations
Dates:
2023/11/15 - 2024/09/30
Student:
Description:
Context of the scientific project
The Spec-LCOCT project is a 3-year collaborative research project gathering 3 academic research laboratories (CRAN, IOGS-LCF and Lip(Sys)2), 1 private company (DAMAE Medical) and 1 University Hospital (CHUSE) to develop spectroscopic line-field confocal optical coherence tomography (LCOCT) for cancer diagnosis improvement in dermatology, funded by the French research national agency (ANR). The project aims at combining the functional information provided by several types of spectroscopies to the morphological information provided by LCOCT in a spectroscopy-enhanced LCOCT device. LCOCT will allow for recording morphological overview images in which points of interest will be subjected to extended characterization using Raman Spectroscopy (RS), AutoFluorescence Spectroscopy (AFS) and Diffuse Reflectance Spectroscopy (DRS). All three spectroscopic modalities will be implemented by an adaptation of the confocal modality of LCOCT and the coupling to a spectrometer. An additional laser source will be required for RS and AFS, while the broadband light emitted by the LCOCT supercontinuum light source will be used for DRS. The acquired data will have to be analyzed to obtain meaningful information. Artificial intelligence (AI) is being increasingly studied for its potential use in medicine, particularly in dermatology. It was demonstrated that AI models, namely classification models based on deep learning methods can achieve the accuracy of board-certified dermatologists for the classification between skin benign and malignant lesions based on dermoscopic image data. In the present project, deep neural network approaches will be implemented to process the acquired data. The expected improvements in skin lesion diagnostic will be studied and quantified in the frame of experimental validations on optical phantoms, ex vivo skin and in clinics.
In this project, our team (CRAN) will bring its expertise in skin AFS and DRS instrumentation development and metrological evaluations [IJ1,4,6,8,13,16,B1], in multidimensional data processing [IJ3-5,7-10,14], in light-tissue interaction modelling [IJ1,2,6,11-12,15,18] as well as in experimental validations on biological tissue-mimicking phantoms but also on preclinical models and in clinics dedicated to skin pathologies [IJ3-5,7,17].

Postdoc objectives and tasks
The 24-month post-doc researcher will join the Spec-LCOCT project's team, within the CRAN "Radiation Interactions with Biological Tissues" group, involving 2 professors (W. Blondel, C. Daul) dealing with numerical estimation of the optical properties of biological tissues from AF and DR spectroscopies as well as multidimensionnal data and image processing, 2 engineers (C. Perrin-Morizet, M. Amouroux) experts in optical instrumentation and experimental protocols on human skin, 1 PhD student (V. Kupriyanov, 2021-24) and a 6month-intern for assisting the experimental protocols.
The recruited post-doc researcher will be involved more specifically in Work packages WP2 (AFS-LCOCT) and WP3 (DRS-LCOCT) of the project to work on the coupling of DRS and AFS modalities to LCOCT, respectively. Both WPs include technological developments and experimental validations tasks. She/he will also develop dedicated algorithms for skin optical properties estimation (absorption and scattering coefficients from DRS and fluorophore concentration from AFS), as well as related signal/image processing, and conduct experimental validations on phantoms and ex vivo samples using CRAN's AF and DR spectroscopic device "Spectrolive".
For sake of confidentiality, complementary detailed information will be provided on demand.
Department(s): 
Biology, Signals and Systems in Cancer and Neuroscience