A system that allows doctors to compare a patient鈥檚 brain scans with a 鈥減ersonalised鈥� atlas created from scans of similar people could reveal subtle and previously hidden abnormalities, say UK doctors.
The prototype Dynamic Brain Atlas was created by a team from King鈥檚 College London, Imperial College London and Oxford University and uses a database of Magnetic Resonance Imaging scans from 200 people. Doctors can select scans from people in the database that most closely match their own patient, in terms of age and sex, for example, to create an averaged 鈥渂rain atlas鈥� for comparison.
Currently, doctors visually compare a patient鈥檚 scans with one or two others. But the comparison could, for example, highlight 鈥渁bnormalities鈥� that are simply the result of age differences.
Advertisement
鈥淭his brain atlas will show the normal range of size and shape of brain structures for a person of the same age, gender and past medical history as the current patient,鈥� says Derek Hill at King鈥檚 College.
鈥淎fter a few seconds, the doctor can see the patient images alongside the brain atlas 鈥� or can see the patient images with features from the atlas overlaid, enabling them to pinpoint the regions of the brain that are abnormal,鈥� he says.
Jo Hajnal at the MRC Clinical Sciences Centre at Imperial College says the approach could have 鈥渉uge implications for patients for whom conventional diagnosis fails鈥�.
Tailored treatment
Very subtle brain abnormalities can underlie disorders such as dementia and psychiatric diseases, Hill says. He thinks the Dynamic Brain Atlas should allow doctors to make very precise diagnoses about the biological bases of different types of dementia, allowing treatment to be more accurately tailored to the patient.
Ideally, thousands of scans of patients should be used as references to create averaged scans, the team says. These could be held in many separate databases in different countries.
Doctors in individual hospitals all over the world would be able to run the system on a standard PC 鈥� but they would require access to substantial computer processing power to deal with the many complex images needed to create a brain atlas for an individual.
Hill believes this will be possible using the computing 鈥淕rids鈥� that many countries are now developing. These are networks of computers, including supercomputers, which individual users can tap into when they need to boost their processing power.
Hill hopes the Dynamic Brain Atlas will be up and running in at least a few hospitals in the UK within a year. The team unveiled the system at the opening of the UK鈥檚 National E-science Centre in Edinburgh.