Description du projet
Un diagnostic tumoral mini-invasif d’une précision maximale
Le diagnostic des tumeurs et la thérapie endoscopiques y afférente souffrent actuellement d’une sensibilité et d’une spécificité limitées, ce qui entraîne des sur-traitements et des sous-traitements, des récidives tumorales, des complications peropératoires et des coûts élevés. C’est pour remédier à ces problèmes que le projet COMBIOSCOPY, financé par l’UE, entend révolutionner l’imagerie endoscopique clinique en combinant la biophotonique et les interventions assistées par ordinateur. L’équipe vise à élaborer des biomarqueurs quantitatifs d’imagerie multimodale qui apportent des informations que l’observation à l’œil nu n’est pas en mesure de fournir. Pour ce faire, les chercheurs combineront des images en lumière blanche avec des images optiques et photo-acoustiques multispectrales afin de reconstruire la surface visible en 3D et les détails anatomiques et fonctionnels de la subsurface. L’objectif ultime du projet est de fournir un diagnostic et une thérapie de haute précision, peu invasifs et peu coûteux. COMBIOSCOPIE est ainsi susceptible de réduire les risques de sur-traitement et de sous-traitement.
Objectif
Key challenges in endoscopic tumor diagnosis and therapy consist of the detection and discrimination of malignant tissue as well as the precise navigation of medical instruments. Currently, a low level of sensitivity and specificity in tumor detection and lack of global orientation lead to both over- and undertreatment, tumor recurrence, intra-operative complications, and high costs. The goal of this multidisciplinary project is to revolutionize clinical endoscopic imaging based on the systematic integration of two new but independant fields of research up until this point: Biophotonics and computer-assisted interventions (COMputational BIOphotonics in endoSCOPY).
For the first time, quantitative multi-modal imaging biomarkers based on structural and functional data are being developed to enhance the physician’s view by providing information that cannot be seen with the naked eye. To this extent, white light images co-registered with multispectral optical and photoacoustic images will be processed in a combined manner to dynamically reconstruct not only the visible surface in 3D but also subsurface anatomical and functional detail such as 3D vessel topology, blood volume and oxygenation. Spatio-temporal registration of multi-modal data acquired before and during the procedure will enable (1) the highly specific local tissue classification and discrimination based on tissue shape, texture, function and radiological contrast imagery as well as (2) global context-aware instrument guidance.
This innovative approach to radiation-free real-time imaging will be implemented and evaluated by means of computer-assisted colonoscopy and laparoscopy. The potential socioeconomic impact of providing high precision minimally-invasive tumor diagnosis and therapy at low cost is extremely high.
Champ scientifique
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringsensorsoptical sensors
- natural sciencesphysical sciencesnuclear physics
- social sciencessociologysocial issuessocial inequalities
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
- natural sciencesphysical sciencesopticsfibre optics
Programme(s)
Thème(s)
Régime de financement
ERC-STG - Starting GrantInstitution d’accueil
69120 Heidelberg
Allemagne