EEG brain trace 'can detect autism'

Post Reply New Topic

A simple brain trace can identify autism in children as young as two years old, scientists believe.

A US team at Boston Children's Hospital say EEG traces, which record electrical brain activity using scalp electrodes, could offer a diagnostic test for this complex condition.

EEG clearly distinguished children with autism from other peers in a trial involving nearly 1,000 children.

Experts say more work is needed to confirm the BMC Medicine study results.

Early detection

There are more than 500,000 people with autism in the UK.

Autism is a spectrum disorder, which means that it is not a single condition and will affect individuals in different ways.

Commonly, people with autism have trouble with social interaction and can appear locked in their own worlds.

It can be a difficult condition to diagnose and can go undetected for years.

The latest study found 33 specific EEG patterns that appeared to be linked to autism.

These patterns consistently spotted autism in children across a range of age groups, spanning from two to 12 years old.

Hallmark brain activity

The researchers repeated their analysis 10 times, splitting up their study group (children with a medical diagnosis of autism and children with no signs of autism) in different ways.

Around 90% of the time, the EEG patterns could correctly detect the children diagnosed with autism.

The team now plan to repeat their study in children with Asperger's syndrome - one particular subset of autism. Typically, people with Asperger's have higher-than-average intelligence and struggle less than people with other types of autism with their speech.

Dr Frank Duffy who is leading the investigation said the work could help determine if Asperger's should be thought of as an entirely separate condition.

And it could point the way to determining if younger siblings of children with autism are likely to develop the same condition themselves.

"It is a great cause of anxiety when an older sibling develops autism.

"EEG might offer a way to check for the same condition in younger siblings in advance of them having symptoms."

EEG could also be used to track what effect different autism treatments are having on the condition, he said.

Caroline Hattersley of The National Autistic Society said: "We welcome any research that may help us to understand autism better and improve diagnosis times for those with the condition.

"In a recent survey we commissioned, 50% of people with autism and their families said it was difficult to get a diagnosis and 55% said the process took too long.

"While further testing of EEG scans is still required, any tools that help identify autism at a younger age could potentially improve a person's quality of life by allowing the right support to be put in place earlier."

Source: http://www.bbc.co.uk/news/health-18577327

I just found this, too, while at Twitter. I was wondering what folks might have to say about it... I guess the question of whether Asperger's really is a form of autism might eventually be settled this way, not by some kind of debate.

I may be taking this a little bleakly. Whatever the answer will be, for as long as the diagnosis of AS exists as an international standard diagnosis and the condition is still going to be often diagnosed without any such future and ("costly") means of physical/neurological verification (including unavailability of those due to too many inexperienced professionals when it comes to ASDs/these specific EEGs), it's unlikely to change anything about the number of misdiagnoses within the spectrum (one type of autism incorrectly diagnosed as another or even non-verbal cases with most other autistic symptoms going undiagnosed completely because of popular stereotypes about the labels). If there's going to be an answer, it would better be put into very practical and easily available use.

this is really neat. i hope they can identify patterns in aspergians also, it could make diagnosis a lot simpler

I concur however this test would cost less than the currant standards in most places I paid out of pocket for an EEG and It cost less than two sessions with the specialist to be diagnosed with or with autism.

Actual science could make the APA's DSM definition definition seem like casting chicken bones.

I wish they had less of a focus on children, and more of a focus on people in general. Adults need a diagnosis as badly as children do, yet adults go completely ignored.

I've been diagnosed as having HFA since I was a young child - so I don't know the issues of not being diagnosed are firsthand, but reading the posts here over the years I have, I see a lot of problems caused by not being diagnosed, and a lot of difficulty for some in getting diagnosed as an adult.

I like that! There would still be the issue of how someone's wiring was affecting their life, however. Other aspects of one's makeup, as well as their environment and personal experiences, would still play a large part in their functionality and other details. The nature/nurture dichotomy is a myth. It's not either/or. So, I suppose the APA might still have something to say, by way of the DSM. I guess we'll have to wait and see how all this goes...

I'd like to know the false-positive rate on this one. If you scan a bunch of NTs, how many of them will seem autistic? What about if you take a bunch of neurodiverse people who aren't autistic--how many of them get mislabeled?

A diagnostic test, to be useful, has to avoid misdiagnosing people, whether by giving them the wrong diagnosis or by giving a diagnosis when they don't need one. If this test can't do that, then it'll be yet one more interesting but useless project.

By the way: Psychology is actual science, and a human observer is a legitimate way of gathering data. When we use tests that don't involve human observers, it's not necessarily better or more accurate. It's simply another perspective on the situation. (And by the way--who d'you think interprets that EEG? Yeah. A human.)

Cool.

If I get diagnosed (one appointment to go, a couple of weeks from now), I plan to start trying to do something more than just gripe on Wrong Planet. Some others who've been diagnosed as adults have written books and stuff, and I'll be thinking about how I want to approach this. I have to wait just a little longer, so that I don't jump too quickly and then feel silly if that's not the diagnosis. But then, if this goes the way it's looking like it might, I'm going to be making some sort of plan of action.

As for children, the more they catch them, the fewer will have to wait until they're adults to get diagnosed. But, yeah, evaluation of adults shouldn't keep involving on-the-spot modifications of tests and questionnaires for little kids. We've found that frustrating in my case, as my girlfriend and sister have gone about trying to help me. And more needs to be understood about masking; compensation; late, eventual learning when it's simply slowed or delayed; and how ASD fits into a picture that includes adult self-awareness and forms of development that wouldn't be present in preschoolers and would be much less likely or more hidden in young kids, such as gender identity and sexual orientation. (Those who followed my story last fall know of how these things became a stumbling block for the psychologist who first evaluated me.) Even continuing education and employment, for those capable of them, can factor in, as an area of interest can appear to simply be someone's job. And all of this is only going to matter to someone who somehow gets clued in that ASD might be an issue for them, so that they can start looking for a place to get assessed. Then they've got to find a place capable of it and willing to do so. That makes the matter a pretty complicated and challenging thing.

Thanks for the words of support, Atomsk!

This is good interesting news, i'll have to research. 33 good number

If you read the post carefully, they said they "split autistics and NTs in various ways". That implies that they DID use NT kids.

On the other hand, what you mentioned about other disorders is a good point. It might be nice to bring groups of people suffering from schizophrenia and depression to see if they will end up testing positive for autism or not.

Regarding "interpreting the test" it is possible to write a comptuer program to do so and that way "human prejudices" wont play any role. I mean, the fact that they found 33 specific patterns would suggeset that it is possible to program them into computer, and most likely thats what they did.