Team Asclepios

Members
Overall Objectives
Scientific Foundations
Software
New Results
Contracts and Grants with Industry
Other Grants and Activities
Dissemination
Bibliography

Section: Contracts and Grants with Industry

Health-e-Child

Participants : Xavier Pennec [ Correspondant ] , Nicholas Ayache, Maxime Sermesant, Hervé Delingette, Stanley Durrleman, Ender Konukoglu.

The European project Health-e-Child (IST 027749, http://www.health-e-child.org/ ), coordinated by Siemens, Germany, aims to create an IT platform to share paediatric knowledge and clinical data based on grid technologies. The project currently brings together eight European countries and intends to integrate heterogeneous biomedical data from three clinical specialities (cardiology, neurology and rheumatology) coming from three paediatric hospitals in Europe (Hôpital Necker in Paris, France, Giannina Gaslini institute in Genoa, Italy, and Great Ormond Street Hospital in London, Great-Britain). This integration should lead to a better understanding of the pathologies studied, and, in the long term, provide real tools to help paediatricians make the right decisions.

Based on our previous works on anatomical statistics and on the modelling of the heart and brain tumours that integrate in particular image and electrophysiological data, the role of the Asclepios team is to model the pathologies that Health-e-child is focussing on. The method is to increase the scope of the integration to more biomedical data (clinical, epidemiological, genetic, etc.) and with practices in place on the different sites. The main difficulty lies in that fact that children's organs are still growing and the pathology interacts with this growth and affects the future evolution of the organs concerned. It is therefore essential to be able to distinguish changes due to pathology from those due to growth. Furthermore, it is also vital that the generic models can be adapted to the specificities of each child. This is why the statistical approach is another important aspect of our modelling, and our participation in the project enables us to access large distributed image data bases to that end. On the other hand, the statistical dimension should make it possible to demonstrate hidden links between different data, for example, between a type of anatomical symptom and a genomic characteristic.


previous
next

Logo Inria