You are here : EN > Research > WP-5 Image-based Simulation and Modeling
-
Partager cette page
WP-5 Image-based Simulation and Modeling
Numerical modeling and simulation are essential disciplines today to address a wide variety of questions in the biomedical sciences, whether they concern the understanding of complex biological systems, the design of high-tech imaging devices or the planning of irradiation therapy treatments. Modeling and simulation play a major role in the exploitation and improvement of existing imaging modalities, as well as in the invention of new imaging concepts (for example, design and development of new instrumentation, testing of new algorithms for reconstruction). The main challenges are to develop and experimentally validate multipurpose simulation tools, which are both accurate and fast, and which allow to address problems of increasing complexity.
Several international initiatives, including the European projects EU-Heart and VPH (Network of Excellence of Virtual Human Physiology) develop methodologies for digital medicine, personalized and predictive. The accuracy of the numerical models provided to simulators is a central point for simulation and modeling. Patient-specific models derived from CT or MRI techniques show increasing applications for simulation in various fields. The very precise control of the dose deposited in the patient during a therapeutic irradiation or during an imaging is also a major public health challenge, as it is studied by several European consortia (ENVISION, ULICE, PARTNER). The intensity of computer work in simulation requires an additional step of abstraction of resources while respecting the applications. Large scale platforms based on the pooling of large groups (GRIDS like EGI, Teragrid, Decrypthon), large data centers (Clouds such as Amazon Web Services, Rackspace) or voluntary PC collections (Destktop computing platforms) are now available to researchers.
In addition, some partners are actively involved in the European HPV Network of Excellence and have expertise in several imaging modalities including MRI, CT, PET, ultrasound imaging and skills to develop future modalities. Compared to other platforms, the main originality lies in the multi-modality criterion, as illustrated by the Virtual Imaging Platform (VIP) project. Image-based models have been used in simulations for various applications, to take into account the effects of postures on the position and shape of the abdominal and thoracic organs.
The partners of this LabEx are also involved in the European collaborative project ENVISION, started in 2010, in which the WP6 is devoted to simulations in hadrontherapy.
Special attention is given to the following two open-source platforms: the Virtual Imaging Platform (VIP) and GATE projects, based on a GEANT4 Monte Carlo toolbox, used by more than 1500 users in the world. Researchers from this WP actively contribute to this platform and to the GATE-lab, an online platform for running GATE on distributed computing resources. Partners also have strong and internationally recognized expertise in high performance computing. This expertise covers, among others, the development of DIET used by the Decrypthon project which was the basis for creating the Sysfera spin off company, parallel and distributed programming model flow planning, cloud optimization and volunteers PC networking, easier access to fences for simulations.
Another objective of this WP is to develop geometrical and biomechanical models of the patient in order to simulate the movement of the organs during the breathing as well as models of the irradiation rooms for the securing of the treatment.
Workpackage 5 follows the idea that, by 2020, advances in modeling and simulation will provide new medical services, in which each patient will bring their own personalized digital avatar built from images harvested in vivo and usable in simulated treatments or imaging for future medical planning. The partners will develop specific tools for multimodal virtual imaging, for dose deposition simulation, for virtual modeling of tissues, organs and tumors, for specific patient modeling, including electrical or biomechanical modeling. In addition, this WP will address high-level computing, as well as data storage and processing issues. Advances in the field of simulation offer new perspectives for the study of biological models, the quantitative evaluation of image analysis methods, the design of new imaging devices or the development of reconstruction algorithms in close relation with WP4.
Several international initiatives, including the European projects EU-Heart and VPH (Network of Excellence of Virtual Human Physiology) develop methodologies for digital medicine, personalized and predictive. The accuracy of the numerical models provided to simulators is a central point for simulation and modeling. Patient-specific models derived from CT or MRI techniques show increasing applications for simulation in various fields. The very precise control of the dose deposited in the patient during a therapeutic irradiation or during an imaging is also a major public health challenge, as it is studied by several European consortia (ENVISION, ULICE, PARTNER). The intensity of computer work in simulation requires an additional step of abstraction of resources while respecting the applications. Large scale platforms based on the pooling of large groups (GRIDS like EGI, Teragrid, Decrypthon), large data centers (Clouds such as Amazon Web Services, Rackspace) or voluntary PC collections (Destktop computing platforms) are now available to researchers.
In addition, some partners are actively involved in the European HPV Network of Excellence and have expertise in several imaging modalities including MRI, CT, PET, ultrasound imaging and skills to develop future modalities. Compared to other platforms, the main originality lies in the multi-modality criterion, as illustrated by the Virtual Imaging Platform (VIP) project. Image-based models have been used in simulations for various applications, to take into account the effects of postures on the position and shape of the abdominal and thoracic organs.
The partners of this LabEx are also involved in the European collaborative project ENVISION, started in 2010, in which the WP6 is devoted to simulations in hadrontherapy.
Special attention is given to the following two open-source platforms: the Virtual Imaging Platform (VIP) and GATE projects, based on a GEANT4 Monte Carlo toolbox, used by more than 1500 users in the world. Researchers from this WP actively contribute to this platform and to the GATE-lab, an online platform for running GATE on distributed computing resources. Partners also have strong and internationally recognized expertise in high performance computing. This expertise covers, among others, the development of DIET used by the Decrypthon project which was the basis for creating the Sysfera spin off company, parallel and distributed programming model flow planning, cloud optimization and volunteers PC networking, easier access to fences for simulations.
Another objective of this WP is to develop geometrical and biomechanical models of the patient in order to simulate the movement of the organs during the breathing as well as models of the irradiation rooms for the securing of the treatment.
Workpackage 5 follows the idea that, by 2020, advances in modeling and simulation will provide new medical services, in which each patient will bring their own personalized digital avatar built from images harvested in vivo and usable in simulated treatments or imaging for future medical planning. The partners will develop specific tools for multimodal virtual imaging, for dose deposition simulation, for virtual modeling of tissues, organs and tumors, for specific patient modeling, including electrical or biomechanical modeling. In addition, this WP will address high-level computing, as well as data storage and processing issues. Advances in the field of simulation offer new perspectives for the study of biological models, the quantitative evaluation of image analysis methods, the design of new imaging devices or the development of reconstruction algorithms in close relation with WP4.
PRIMES thesis on WP5 themes:
- Anchen CHAI 2015-2018 - Simulation réaliste de l'exécution des applications déployées sur des systèmes distribués avec un focus sur l'amélioration de la gestion des fichiers
- Benjamin DELPUECH 2016-2019 - Simulation de la résistance du tibia de souris avec et sans tumeur osseuse
- Brent HUISMAN 2013-2016 - Accelerated clinical prompt gamma simulations for proton therapy
- Karolina GOLEC 2014-2017 - Hybrid 3D Mass Spring System for Soft Tissue Simulation
- Loriane WEBER 2013-2016 - Iterative tomographic X-Ray phase reconstruction
- Louis RIGLET 2019-2022 - Prothèse totale de hanche double mobilité : mobilité et stabilité
- Marc GARDEGARONT 2020-2023 - Évaluation in vivo du risque de fracture de fémurs métastatiques
- Odran PIVOT 2016-2019 - Scatter correction for spectral computed tomography
- Victor LEVRAGUE 2021 - 2024 - Modélisation biophysique pour les thérapies ciblées impliquant l’émission d’ions de basse énergie
- Yasmine ALI 2018 - 2021 - Estimation de la dose biologique en hadronthérapie avec la plateforme de simulation Monte Carlo GATE
- Yazid TOUILEB 2014-2019 - Four-dimensional dose calculation using deformable tetrahedral geometries for hadron therapy
WP-5 leaders
Lydia MAIGNE, LPC Clermont laboratory
David MITTON, LBMC laboratory Lyon