Speakers
Description
Keywords
Surveillance network, cancer, databases federation, epidemiology, data exchange
Impact
First, the objective was to offer computerized anatomical pathology reports access for cancer screening associations by linking in a grid infrastructure data holders and users. After that, if such a network can federate the entire regional anatomical pathology databases, it can easily be fitted to general public health. That’s why regional and national French institutes of public health are really interested and now involved in the project which would constitute a new tool to perform accurate epidemiologic statistics.
The development will start in the beginning of 2009 in collaboration with Maat-G and a prototype will be released in June. The security will be ensured by the grid framework, health-professional smartcard certificates for authentication, data anonymisation for each data movement and a distributed-based way to identify a patient which satisfies legal requirements in terms of data privacy.
URL for further information
clrwww.in2p3.fr
Conclusions and Future Work
This project aims at proving the feasibility and the reliability of a grid-based surveillance network in a real application with real confidential data using grid technology developed by EGEE and AuverGrid projects. The network resources are now enough to create a fully-connected network exchanging data.
If the first part succeeds, a lot of perspectives for the future will be opened: integration of other sources like images or adding users of the surveillance network focusing on epidemiology.
Detailed analysis
In France, the national program for early diagnosis of breast cancer is operated by associations which are in charge of encouraging every three years women above 50 years old to be screened by mammography. If the result is positive, a biopsy of the suspect tissue is forwarded to anatomical pathology laboratories. Because of this, the entire diagnoses of cancers are available there through anatomical reports.
The main requirement for data holders, i.e. anatomical-pathology laboratories or radiology services, is to ensure ownership of their data. The grid technology is particularly well fitted for such an approach: a grid can federate databases and provide a secured framework where patient data are kept in the laboratory.
Using grid security mechanisms, some identified users would be able to query and retrieve data from owners that agreed and trusted beforehand this customer. Therefore, providers keep the complete control of their data.