Sophisticated brain imaging can be used to predict with 90% accuracy which teenagers with dyslexia would improve reading skills over time, according to a new study conducted by researchers at the Stanford University School of Medicine.

The researchers conducted a study to determine whether neuroimaging could predict reading improvement, and enrolled 25 children with dyslexia and 25 children with typical reading skills, who were aged around 14 years.

The researchers assessed their reading with standardised tests, and then conducted two types of imaging – functional magnetic resonance imaging and diffusion tensor imaging as children performed reading tasks.

A follow-up study after two-and-a-half years was conducted to determine which brain image taken at baseline would predict the reading improvement in the child.

The study revealed that children with dyslexia who at baseline showed greater activation in the right inferior frontal gyrus during a specific task and whose white matter connected to this right frontal region was better organised, showed a greater improvement in reading during the study period.

In addition, by looking at patterns of activation across the whole brain, researchers were able to very accurately predict future reading gains in the children with dyslexia.

The researchers noted that fMRI is typically viewed as a research tool that has little practical implication for an individual with dyslexia.

"Yet these findings suggest that, after additional study, brain imaging could be used as a prognostic tool to predict reading improvement in dyslexic children," the researchers said.