How can a neurosurgeon be sure that he removed the totality of the tumour from his patient's brain, while preserving healthy neighbouring tissues ? To help the practitioner, researchers and physicians collaborate to develop a camera able to reliably distinguish healthy cells from diseased cells. To do this, it will analyse the fluroescence naturally emitted by the tissue, rich in information, and compare the result with a database before giving its final result: healthy tissue or tumour tissue.
In 2017, to feed this database, researchers come to, among others, Synchrotron SOLEIL on the DISCO beamline, to establish the characeristics of the different types of cells in the ultraviolet.
Acknowledgments:
- Institut National de la Santé et de la Recherche Médicale, l'AAP (2012-2014-2016/ MEVO & IMOP) - PLAN CANCER
- Agence Nationale pour la Recherche au titre du programme « Investissements d’avenir » - FLI
- Mission pour l’interdisciplinarité « l’instrumentation aux limites » - CNRS
- Plateforme « PIMPA », Paris Sud
- Université Paris Diderot, Sorbonne Paris Cité
- Laboratoire IMNC-UMR 8165-CNRS/IN2P3, Université Paris-Saclay
Thanks to Pr. Bertrand Devaux who has supported this project since the beginning, and Pascale Varlet, neuropathologist.
This video was partly made in the Sainte-Anne Hospital Centre (Paris).
Autumn 2024: a look back at this research
Since 2012, physicist Darine Abi Haidar has dedicated her research to improving the surgical procedure for removing brain tumors. This is an extremely complex task because these tumors are "infiltrating," meaning they grow insidiously within the surrounding tissues of the initial site. It is only through an anatomical pathology analysis, performed after the surgery, that doctors can determine if all the diseased tissue has been successfully removed.
Darine has set out to create an intraoperative diagnostic tool that can be used during the surgery itself. This tool is essentially a kind of probe that the surgeon can hold and pass over the brain, providing clear signals indicating whether the tissue is healthy or not. To achieve this, the optics specialist leverages the natural fluorescence of tissues. When illuminated with certain lights, the tissues become "excited" and emit another light, revealing some of their characteristics. Darine first chose to observe and understand this fluorescence using numerous tissue samples, working in the infrared spectrum on the PIMPA platform, and in the ultraviolet spectrum at the synchrotron SOLEIL on the DISCO beamline.
"My collaboration with SOLEIL has enabled several publications," Darine explains. "The analysis on DISCO has been particularly useful in determining the grade (severity) assigned to meningiomas, and also in investigating metastatic tumors to trace their origin in the body."
The tool Darine is developing involves three key components: the database, the instrumentation, and artificial intelligence. For the database, no fewer than 300 brain tissue samples collected from hospitals have been meticulously analyzed using fluorescence techniques. The instrumentation involves designing the probe, equipped with an optical fiber. The signal captured is then analyzed by a program trained using the database, forming the "artificial intelligence" layer. This software aspect employs cutting-edge techniques (machine learning, deep learning).
The human-machine interface also needs to be considered—this is the system that allows the surgeon to view the information in a clear and simple manner. It is even planned that the same information will be available in real-time to a pathologist, enabling a cross-analysis of expertise—once again, during the surgery. One can imagine that certain information, such as colors representing different types of tissue, could be superimposed onto the brain's image on a screen.
The ambitious project (MITA-OPALIS) continues to grow. It collaborates not only with Sainte-Anne Hospital but also with La Riboisière in Paris and Henri Mondor in Créteil. More recently, the project has expanded to investigate other types of tumors, including those in urology (bladder), ENT, and gynecology (breast). "The urological field is particularly interesting because there are many applications, and the tumors are easier to delineate than those in the brain," Darine comments.
Patents—currently confidential, of course—are in the process of maturing and will soon be filed, as Darine enthusiastically hopes. The logical outcome of this would be the potential creation of a startup.
Related publications
Mehidine, H., Refregiers, M., Jamme, F., Varlet, P., Juchaux, M., Devaux, B., Haidar, D.A. "Molecular changes tracking through multiscale fluorescence microscopy differentiate Meningioma grades and non-tumoral brain tissues" Scientific Reports., 11: art.n° 3816. (2021).
Mehidine, H., Chalumeau, A., Poulon, F., Jamme, F., Varlet, P., Devaux, B., Refregiers, M., Abi Haidar, D. "Optical Signatures Derived From Deep UV to NIR Excitation Discriminates Healthy Samples From Low and High Grades Glioma" Scientific Reports., 9: art.n° 8786. (2019).
Poulon, F., Chalumeau, A., Jamme, F., Pallud, J., Varlet, P., Mehidine, H., Juchaux, M., Devaux, B., Refregiers, M., Abi Haidar, D. "Multimodal Analysis of Central Nervous System Tumor Tissue Endogenous Fluorescence With Multiscale Excitation" Frontiers in Physics., 6: art.n° 109. (2018).