Aller au contenu principal


IMPACT: Images and Models for the Planning and AssistanCe to surgery and Therapy

Team leaders: Pascal HAIGRON (PU) & Renaud DE CREVOISIER (PU-PH)


Areas: Image Guided Therapies - Computer Assisted Interventions

  • Cardiovascular therapy

  • External radiation therapy

Objective: Optimization and personalization of the interventional strategy



  • Moving / deformable anatomical structures

  • Pathophysiological observations: pre-/per-/post-treatment,  multi-modal, multi-scale


The research of IMPACT team deals with computer assisted and image guided therapy. The aim is to contribute to the development of new minimally-invasive procedures for the treatment of cardiovascular and cancer diseases. These interventional techniques are permanently evolving and are to become increasingly targeted and accurate thanks to the coupling of multi-modal and complementary observation-action processes. The challenge is to secure, optimize and personalize the therapeutic strategy within a context of soft tissues, where organs and lesions deform or move during the treatment. Our work is mainly oriented towards therapy delivery which requires accurate monitoring and control of interactions between the implantable or treatment devices and the pathological tissues. The aim is to contribute to the elaboration of decision support solutions integrating N-dimensional, anatomo-functional, and multiscale data (observations, predictive models), but also, with the improvement of treatment targeting, to the better understanding of the pathophysiological mechanisms underpinning the response to therapy. This implies the development and validation of new algorithmic tools operating in a translational approach and in close relationship with the interventional practices in which they fit. In this context our efforts mainly focus on: (i) precise description and localization of anatomo-pathological targets, (ii) prediction of therapy outcome and (iii) adaptation and personalization of interventional procedures.

Key words:

Image Guided Therapy, Computer-Assisted Interventions, Cardiovascular Therapy, Radiation Therapy, IGRT, Minimally-Invasive Surgery, Interventional Imaging, Multimodal Data Fusion, Augmented Endovascular Navigation, Patient-Specific Simulation, Predictive Modeling. LTSI_image_IMPACT2_0.jpgPatient-specific real-virtual cooperation: an integrative approach for treatment personalization [Haigron et al, IEEE Eng. in Med. & Biol. Mag. 2009;28(4):96-8]

Computer assisted medical and surgical interventions (with either passive, synergistic or active guidance) can largely take benefit of virtual/mixed reality like approaches. Our approach was very early structured within a real/virtual cooperation framework with a generic scope, i.e. transposable to the different physical (or energy-based) therapeutic principles which interest us, whatever their mode of application is (external, interstitial, intra-luminal/cavitary).

Such a scheme involves decision-making loops (critical tasks performed by the user), as well as reactive loops (computational tasks performed by the machine). By integrating more information from heterogeneous data and predictive modelling, our aim is to contribute to the elaboration of new knoweldge and solution to support interventional decision making, which is spread over different timescales and imposes the combination of very-short-term (intra-operative planning and action), short-term (pre-operative planning), and long-term (follow-up, recommendations) analyses.

In this context, our work more specifically addresses the issues of: