Jul 10 2006
Software that gives doctors much-needed help in the diagnosis and treatment of brain tumours is being developed by an international consortium of scientists and medics.
The technology will give medical staff access to the results of hi-tech scans and laboratory tests gathered from across Europe and will enable them to respond more effectively to new cases of this disease which is notoriously hard to treat. The unprecedented access which the HealthAgents initiative affords will be of particular benefit to the many children and young adults who develop these tumours.
The 4.5-million Euro project, funded by the EU, involves scientists and medics from the Universities of Birmingham, Edinburgh, Southampton, Valencia, Barcelona and Leuven as well as three international companies, Microart, PQE, and ITACA.
Brain cancer has a relatively high incidence among children and young adults and the rates of mortality and morbidity (level of debilitation) remain high making this a particularly important healthcare initiative.
Diagnosis at present requires a surgical biopsy which has a risk attached to it and developing scans which provide a non-invasive diagnosis would be a major advance. More important still would be the development of scans which can tell doctors how well a tumour will respond to particular treatments allowing treatment to be tailored to the individual
Translating advances in technology into improved patient care is a major challenge. The use of new techniques by doctors is often impeded by difficulty gaining access to the experience of expert centres.
HealthAgents software will allow doctors to make more informed decisions because it gives them access to the results from hundreds of cases drawn from more than 50 hospitals across Europe, including the UK Children's Cancer Study Group (UKCCSG). All data will be stored anonymously, and securely.
Central to the HealthAgents approach is the use of so-called agent technology –a series of software models that encourage a more consensual approach to diagnosis. The software will enable doctors to compare their own cases with the results of cases from both their own and other hospitals.
The technology will allow doctors to input data from hi-tech scans of their own patients into the software package, which uses pattern recognition techniques to classify the scan results with those of similar cases from all participating hospitals. The software then suggests a series of diagnoses which the doctor considers when determining treatment.
According to Dr Andrew Peet, a member of the UKCCSG and clinical advisor to HealthAgents, the project gives doctors access to a range of securely held results that would simply not have been previously available.
He added: "The consortium encompasses a wide range of talent from universities, companies and hospitals. Their expertise of the consortium will provide the foundations for the first EU grid for brain tumour diagnosis and prognosis. We will profit from a world-class data exchange network to address one of the most pernicious diseases of our time."
For further information, please contact:
Horacio Gonzalez-Velez, University of Edinburgh
Tel: +44- (0) 131 650 5152 Email: h.gv (at) ed.ac.uk
Ronald Kerr, University of Edinburgh
Tel: +44- (0) 131 650 9547 Email: Ronald.Kerr (at) ed.ac.uk
AI models can be trained to distinguish brain tumors from healthy tissue