Brain imaging system reveals details of autism development

Scientists have shown for the first time that measuring brain activity in babies in their first six months of life may help reveal differences between the brain responses of infants at-risk for autism and infants with no family history of autism. The findings suggest that direct brain measures might help to predict the future development of autism symptoms in infants as young as four to six months, although the researchers emphasize that the study is only a first step towards earlier diagnosis of the disorder. The measurements had not been possible until now because no existing imaging system was suitable for this delicate task, but were achieved using a special near-infrared ‘helmet’. See a video about the work:

The research, which was funded by the UK Medical Research Council, The Simons Foundation and Autistica, was conducted at the Centre for Brain and Cognitive Development, Birkbeck, University of London using the device developed by researchers at UCL Medical Physics & Bioengineering. Results are published in the 13th March issue of the Proceedings of the Royal Society B as the paper “Reduced neural sensitivity to social stimuli in infants at risk for autism”, available to read freely online.

In their first six months of life, babies who have an older brother or sister with autism show different brain responses compared with a group of babies with no autism in the family. The activation of the brain can be measured by changes in the way near infra-red light reflects off areas of the brain depending on the level of oxygen used there. Differences can be seen between the babies viewing socially interesting videos of, for example, actors playing ‘Peek-a-boo’, and listening to vocal sounds such as laughter and yawning, compared with sounds such as running water and toys rattling.

Dr Sarah Lloyd Fox, who led the study, explained: “Our findings demonstrate for the first time that direct measures of brain functioning during the first six months of life may help further our understanding of the development of autism. Future work will determine whether these differences in brain responses to socially relevant information are associated with later autism or the broader autism phenotype, which is sometimes seen in unaffected family members.”

The behaviours characteristic of autism emerge over the first few years of life and firm diagnoses are now made in children only after the age of two. As a result, the vast majority of research on autism has necessarily concentrated on children two and up, who have already been diagnosed.

“We still know very little about the earliest appearing symptoms and warning signs,” Lloyd-Fox said.

To find out more, the team looked to four-to-six-month-old babies at greater risk of developing autism because they had an older brother or sister with the condition. The researchers used an optical imaging system which had been custom designed and built by engineers from the Biomedical Optics Research Laboratory at UCL Engineering. Optical sensors were placed on the baby’s head to register brain activity while they viewed videos of human actions (such as Peek-a-boo or Incy Wincy Spider) or listened to sounds such as laughter, yawning, running water and toys rattling.

“At this age, no behavioural markers of autism are yet evident, and so measurements of brain function may be a more sensitive indicator of risk. The earlier that we can measure infants’ responses, the clearer an idea we can develop of how genes and the environment might be interacting, and this will help us to develop interventions which could support typical brain development.” Lloyd-Fox said.

The UCL near-infrared brain imaging device is a cap that can be put on an infants head while it is held by a carer.

The UCL near-infrared brain imaging device is also finding applications in developing countries, due to its portability.

Lloyd-Fox explains “It is important to note, however, that individual babies did not all show the same pattern of brain responses. It is paramount that we revisit these findings when the babies are over two years of age and can be assessed for a diagnosis of autism. Future work will determine whether these differences in brain responses to socially relevant information are associated with later autism or the broader autism phenotype, which is sometimes seen in unaffected family members.” The method will require further refinement, most likely in combination with other factors, to form the basis of a predictor accurate enough for clinical use in the general population.

Professor Clare Elwell (UCL Medical Physics & Bioengineering), whose group developed the optical sensors that made this work possible, said: “It’s really satisfying to produce work which enables important developments in other disciplines. The gentle nature and portability of our technique of near infra-red measurement means that we have been able to find uses for it in athletics, developmental studies such as this, and are currently exploring applications it could have in developing countries.”

The research has been funded by the UK Medical Research Council, Simons Foundation and the BASIS funding consortium led by Autistica, a charity seeking to fund biomedical research to bring benefits to individuals and families affected by autism spectrum disorders.

Read more