The analysis of brain images may improve diagnosis and treatment of mental health problems

Article originally written in November 2014

Peter Liddle and his research group aim to take scientific information about how the brain works and use it to diagnose mental illnesses and suggest the best treatment in clinical practice.

Psychiatry expert Peter Liddle is the director of the new Centre for Translational Neuroimaging at the Institute of Mental Health, in Nottingham, launched on the 30 October. His multidisciplinary team has five permanent researchers and around 20 contributors.

Sophisticated technologies have been used for 20 years or more to produce images of both the structure and the functioning of the human brain. But the brain is a very complicated organ, and one that varies widely between people. Liddle says that this complexity and variability make it difficult to translate and use in practice the information we already have.

Institute of Mental Health, Nottingham
Institute of Mental Health, in Nottingham, the home of the Centre for Translational Neuroimaging.


To add to this, quite subtle things about the mind are affected by mental disorders, although they may be very disruptive.

Liddle says: “Even a person with schizophrenia, most of what they say makes sense, most of what they feel are emotions that any other person would have in similar circumstances”.

It is no surprise then, that the range of measurements made in the brains of healthy people and of patients with anomalies can overlap rather significantly. This makes it difficult to use that information to reliably diagnose individuals.

“At the moment we can only draw statistical conclusions for groups of people,” Liddle says, but “the key thing is that you have to be able to draw useful conclusions about one person.”

“Where I think our research group is unique is in saying that we should put a lot of effort into getting improved predictions in the next couple of years.”

It is by adding more and more information that Peter Liddle and his group of researchers expect to make better predictions for individual cases. Typically, any one image of the brain may show 50, 000 measurements. That is enough to make confident predictions once they have developed the mathematical procedures to analyse all that data.

“I was appointed here in Nottingham in 2001 to really get this thing going, but it is only really in the past year or so that I am now fairly confident that we can really use this in practice,” Liddle says.

The Centre for Translational Neuroimaging is so far focused on three mental health conditions. “Schizophrenia is the most seriously disabling of the mental disorders that affect young and middle aged adults,” Liddle says. “Roughly speaking, one in a hundred people sometime in their life will have schizophrenia,” he adds.

Liddle also considers Attention Deficit/Hyperactivity Disorder (ADHD) an important problem due to its effects on childhood development.

His team also studies depression because it is very common and the existing treatments work for some people but not for others or have no long-term effect.

“Where I think our research group is unique is in saying that we should put a lot of effort into getting improved predictions in the next couple of years, whereas other people would be saying ʻLetʼs really understand the whole thing betterʼ”. Liddle says that more 30 years are needed to be able to say for certain that a given person has schizophrenia. But if in three yearsʼ time his team could prove that imaging information could be used on an individual to predict the evolution and the effectiveness of treatment in early psychosis, he would consider that to be a wonderful achievement.


Have your say