Scientific

• Improve the classification of brain tumours by using a distributed network of local databases or Data Marts.
• Develop new pattern recognition methods for a distributed classification and analysis of HR MAS and microarray data.
• Define a method to check the quality and usability of a new candidate local database with a set of new cases, based on a quality score.
• Compile evaluate and use parameters to audit classifiers and improve them periodically (First level of “self learning.”)
• Create a “trust” framework to increase the amount of collaborating centres that after classifying some new cases will possibly provide new brain tumour cases for the distributed data warehouse
• Publish scientific articles featuring most relevant achievements.

Technological

• Build the world’s largest distributed data warehouse of brain tumour cases data.
• Set up a multi-agent system based on a distributed architecture.
• Design a new ontology of brain tumour data.
• Define a process to convert brain tumour data.
• Devise a new data exchange protocol, to send and retrieve brain tumour data through a multi-agent based network.
• Develop tumour classifiers using the distributed data warehouse.
• Compare and analyse similitude and differences between Data Marts.
• Formulate a protocol to produce improved classifiers from new data.
• Enforce security policies (anonymity, authentication, authorisation, and consistency) of all information system wide.