Forgot your password?
typodupeerror
Medicine Science Technology

Brain Scan Can Detect Autism In Infants 166

Posted by samzenpus
from the early-check dept.
kkleiner writes "A new study shows that brain scans can detect autism in children as young as 6 months old. Researchers at University of North Carolina's Institute for Developmental Disabilities imaged the brains of 92 children who were at high risk for autism. Scans were performed when the children were 6 months, 1-year, and 2-years old. At 2 years, the age when children are typically diagnosed, 30 percent of the children were found to have autism. The researchers then compared the brain images of the autistic children with the others. They saw differences in the brain's white matter, the axon-laden pathways that transmit electrical signals to distant parts of the brain. Of the 15 pathways analyzed, 12 were significantly different between autistic and non-autistic children."
This discussion has been archived. No new comments can be posted.

Brain Scan Can Detect Autism In Infants

Comments Filter:
  • by mvdw (613057) on Thursday February 23, 2012 @08:19PM (#39142471) Homepage
    Immunisation caused autism. Science FTW!
    • Beat me to it!

      • If I can recall correctly, brain scan does introduces radiation into the brain

        What are they thinking?

        Scanning the little brains of little babies will only do more harm than good to those babies

        I do not object of performing brain scan on babies - but on those babies who may have cancer or other fatal illness in order to diagnose their condition

        Autism may be serious, but at least, it's not a fatal sickness

        • by JoshuaZ (1134087) on Thursday February 23, 2012 @08:42PM (#39142661) Homepage

          If I can recall correctly, brain scan does introduces radiation into the brain

          Depends on the type. PET scans do have some radiation. The study in question - http://neuro.psychiatryonline.org/article.aspx?articleid=668180&RelatedWidgetArticles=true [psychiatryonline.org] used diffusion tensor imaging which is a variant of MRI, which uses strong magnetic fields and does't produce any radiation. The technique is essentially harmless.

          • by jamesh (87723) on Thursday February 23, 2012 @08:51PM (#39142717)

            The technique is essentially harmless.

            Except that 30% of those scanned in the study now have autism... coincidence? ;)

          • by quenda (644621)

            But even weak magnets have a biological effect. They must do - magnet therapy is a billion dollar industry.
            If a $2 ferro-magnet can cure cancer, the massive field of an MRI must be able to cause disease. This relationship is proven by the homeopathic principle.

            • Well in all fairness there is:
              http://en.wikipedia.org/wiki/Transcranial_magnetic_stimulation [wikipedia.org]

              Single or paired pulse TMS causes neurons in the neocortex under the site of stimulation to depolarize and discharge an action potential. If used in the primary motor cortex, it produces muscle activity referred to as a motor evoked potential (MEP) which can be recorded on electromyography. If used on the occipital cortex, 'phosphenes' (flashes of light) might be perceived by the subject. In most other areas of the cortex, the participant does not consciously experience any effect, but his or her behaviour may be slightly altered (e.g. slower reaction time on a cognitive task), or changes in brain activity may be detected using sensing equipment.

        • by tgibbs (83782)

          No, diffusion tensor imaging is done with magnetism. No radiation is required and there is no risk of cancer or other fatal illness.

    • by MightyYar (622222)

      These infants all had their entire 2-year regimen early :)

    • Re: (Score:3, Funny)

      by linatux (63153)

      Brain scans caused autism in 30% of the children scanned!

    • Don't get your kids vaccines. In the coming resurgance of polio and measles I get the last laugh.
  • by ackthpt (218170) on Thursday February 23, 2012 @08:20PM (#39142479) Homepage Journal

    It's like a smoke detector which is telling you your house is already on fire, but can do nothing about putting it out or preventing it from happening.

    Might be useful in some way for consulting with parents, but a knife-edge decision there, to decide whether to tell people their child might develop Autism, with the possibility you culd be wrong. While it is interesting information the practice side of how to use it raises some questions.

    • by LehiNephi (695428) on Thursday February 23, 2012 @08:27PM (#39142539) Journal
      There are a couple questions I have about the study:

      1) How close was the correlation between the odd brain scans and incidence of autism? what was the error rate?
      2) Can the brain scan be used to predict autism, rather than as a 20/20 hindsight study?
      3) If the test does reliably predict autism, how practical is it to put it into use? I can't imagine it would make much sense to do it to every 6-month-old--the cost would be prohibitive, particularly since it's a relatively small percentage of the population which are autistic. Would this become something that is only used for children who have a high risk of developing autism?

      /parent of an autistic son
      • by ianbean (525407) on Thursday February 23, 2012 @08:38PM (#39142627)
        1) The original research article is open access so you can take a look for yourself. The differences were statistically significant between the two groups (ASD-positive versus -negative) but there isn't enough data to estimate a sensitivity or specificity for using this as a test. 2) It could be a diagnostic test. In theory that would allow potential treatments to begin before the behavioral aspects of ASD manifest (which is around 2yo correct?)
        • by Shavano (2541114)

          That presumes there is effective treatment for autism in infants, which there is not. The best you can hope for is to warn the parents what they may be facing.

          • by tgibbs (83782) on Friday February 24, 2012 @01:26AM (#39144319)

            Actually, nobody knows whether there is effective treatment for autism in infants. Until autism can be reliably diagnosed in infants, there is no way to test whether any kind of early intervention will improve outcome.

          • Actually, there is an effective, although very limited, treatment for autism. It has been discovered that large doses of B vitamin (I forget which one) reduces the symptoms of autism significantly. It would be interesting to see what impact starting a B vitamin regimen on infants who show positive for autism via this scan has on their eventual behavior (and whether it impacts the way the scan comes out later).
            There were two hypothesis developed from the study that showed that a B vitamin regimen reduced sy
      • by Anonymous Coward

        Studying the brain development of infants, who later are diagnosed as having autism, provides important information as to the root causes of autistic. And understanding the root cause is important in creating effective treatments.

      • by Fned (43219)

        There are a couple questions I have about the study:

        Those sound less like questions about this study and more like questions about the next study that they haven't done yet...

      • by icebike (68054) *

        3) If the test does reliably predict autism, how practical is it to put it into use? I can't imagine it would make much sense to do it to every 6-month-old--the cost would be prohibitive, particularly since it's a relatively small percentage of the population which are autistic. Would this become something that is only used for children who have a high risk of developing autism?

        Unless or Until there is a early intervention/prevention treatment, finding these children early seems expensive and not that helpful.

        AutismSpeaks [autismspeaks.org] is big on therapy but statistical evaluation of success seems pretty spotty at best, since no two kids are the same, and none of these seem appropriate for autism at 6 months to 1-year, which scans might detect.

        Since normal detection typically occurs at 2 years, therapy has at best 12 to 18 months additional time to "work" if scans are performed. But all the th

      • From TFA

        At 2 years, the age when children are typically diagnosed, 30 percent of the children were found to have autism. The researchers then compared the brain images of the autistic children with the others. They saw differences in the brain's white matter...

        They need to do another experiment. It's not right to look at the scans AFTER you know which ones have a problem. Otherwise you start to "see things" to support the conclusion. They don't really need to re-scan, they could just show all the scans to

        • by LurkerXXX (667952)

          It IS right to look at the scans AFTER you know which ones have a problem if your goal is to identify differences. An eventual later goal would be to prove that the differences are definitive and useful for diagnosis. That's not what these researchers were doing. RTFA. They don't claim that this IS diagnostic, but that their research results found differences that are interesting, and might lead down the road to predictive tests. They make no claim that it is a predictive test.

          • by metacell (523607)

            I haven't checked if the study is double-blind, but if it isn't, it doesn't even prove the differences exist. It's very easy to see differences when you expect them to be there.

      • by slew (2918) on Thursday February 23, 2012 @10:07PM (#39143301)

        FWIW, this seems to be just the latest in research that has been coming out in the last few years that seem to add to the evidence that autism onset is really early or preterm.

        For instance, that this paper that came out in 2005 [ucsd.edu] attempted to more systematically document the early signs of autism by using a longitudinal study which comparing a set of high risk infants (who had an older sibling diagnosed w/ some ASD), with a control set at 6 months, 12 months, and 24 months.

        Of course today, autism is defined in terms of behavior. The key is the origin or eitology of autism. Is this odd brain scan an indicator of the origin, or is it something else and this is merely correlated. Nobody knows.

        However, we do seem to have strong indication of certain specific chromosomal problems that can lead to autistic behavior: FragileX, Rett Syndrome, and Tuberous Sclerosis. Fragile X and Rett's are X chromosome related. This is suspiciously related to the observation that incidence of ASD are higher in boys and boys only have 1 "X" chromosome. TS is not X related, but can cause calcium deposits to develop in the brain or in some cases tuber-like growths in the brain (as special case, since tuber-like growths occur all over the body w/ this condition).

        Because of the accumulated research, many people speculate that there are actually many pathways to autistic behaviour. You might even think of autism as a symptom of many diseases and conditions (like a cough is a symptom of many diseases), so many experts are not optimistic that will be a "cure" for autism, but perhaps in the future there will be a way in some situations to blunt the symptoms early enough to avoid many of the problems in many of the common cases.

        Reality is often much more complicated that the headlines.

      • by tobiah (308208)

        2) Autism is a very poorly defined disease, and so probably encompasses many. I seriously doubt a region-specific white-matter analysis of an MRI image will prove useful in predicting it. Perhaps a subset of the patients have a white-matter cause of their disorder, but it would still be very hard to detect. The inspiration for this approach was the similar test done (usually with ultrasound) for Down's, which is a much more clearly defined disorder. But even with that test there are plenty of false positive

    • by MightyYar (622222) on Thursday February 23, 2012 @08:36PM (#39142609)

      It's like a smoke detector which is telling you your house is already on fire, but can do nothing about putting it out or preventing it from happening.

      On the other hand, you can get a 2-year jump-start on locking the kid in the attic.

    • I agree that earlier diagnosis might not have much value beyond consultation, and in fact these brain scans might do more harm than good along that line. But the research might yield clues into what causes autism and, although probably too optimistic, maybe even help lead us to a method of preventing the disorder.
    • by tomhath (637240)
      Not really. There's a reasonable chance that the development of autism could be avoided or minimized if we understand what causes it. Many genetic diseases can be treated (e.g. MSUD [nih.gov], catch it in time and the child can live a fairly normal life).
    • by Tropaios (244000) <tropaios@yahooREDHAT.com minus distro> on Thursday February 23, 2012 @09:24PM (#39142973)

      Phenylketonuria and celiacs are both very common with autism. By carefully controlling the diet at a younger age many behaviors might be curbed or avoided altogether. With this as with all medical conditions, the earlier you know the better.

    • by rrohbeck (944847)

      It's being narrowed down to genetics so hopefully they can identify a gene combination sooner or later. Then potential parents could have a test.

    • Re: (Score:2, Informative)

      by Anonymous Coward

      Our son was diagnosed with autism at just under 2 years old and put into early intervention preschool right when he turned 3. Between that and the things we've done at home, we now have a 9 year old he is doing well in a regular ed class with an aide and who will probably be able to graduate high school and get a job and hopefully live on his own some day.

      If we would have known at 6 months old we could have gotten started earlier and he would be miles ahead of where he is now. The things that kids learn bet

    • by JoeMerchant (803320) on Friday February 24, 2012 @01:15AM (#39144275)

      In 2003, they would not even attempt a diagnosis of Autism before age 3, even though all the research was pointing toward early intervention with socialization therapy as highly effective at mitigating the worst of the debilitating effects of ASD.

      Getting this diagnosis at 6 months can kick off a course of ABA and socialization focus at a time when it matters most to the child's future.

      I just spoke with a surgeon this morning, married to another surgeon, who have a 16 year old son with straight As and no concept of what a friend is - if they had noticed earlier and done something about it, he might (according to the literature) have more of a social life at this point. Maybe he's happy the way he is, but at some point, social withdrawal has a negative impact on the ability to pro-create, and even though children have a negative correlation to happiness, it is sad to not be able to have them if you really want to.

    • The earlier an autistic child starts therapy the better the chances are that s/he will do better. Better diagnostic techniques that let parents know about autism in their children sooner is valuable.

  • Hep A vaccinations aren't given until after one year of age.
  • by Anonymous Coward

    If the brain is showing abnormality at birth or close to it then we can at least rule out post-natal causes (e.g vaccines). Perhaps epidemiologists should start looking into diet deficiencies or exposure to toxins during pregnancy.

    • by fiannaFailMan (702447) on Thursday February 23, 2012 @08:30PM (#39142567) Journal

      If the brain is showing abnormality at birth or close to it then we can at least rule out post-natal causes (e.g vaccines).

      Well they already have been, but maybe this'll convince the Oprah/Jenny McCarthy/Jim Carey crowd.

      • by nedlohs (1335013) on Thursday February 23, 2012 @08:51PM (#39142715)

        Not a chance, they aren't being rational to start with so evidence and facts are irrelevant.

        • by rrohbeck (944847)

          Those evil scientists surely faked their data because they were paid by the medial-industrial complex! Where is autismgate?

      • by Anonymous Coward

        If the brain is showing abnormality at birth or close to it then we can at least rule out post-natal causes (e.g vaccines).

        Well they already have been, but maybe this'll convince the Oprah/Jenny McCarthy/Jim Carey crowd.

        I don't think you can do anything to convince them. About all you can do is just ignore them enough until they cease to be as vocal about this topic and until they find some other cause célèbre to occupy their time.

      • Ha, no. For that we must resort to pointed sticks, like with all blatantly illegitimate vested interests and their detritus.
  • by Martin Blank (154261) on Thursday February 23, 2012 @08:32PM (#39142579) Journal

    The study results do not detect autism (which most people will read as a diagnosis). When looking back over the scans, differences from normal development could be seen starting as early as six months, but this is retrospective after it was diagnosed at the common ages. This study may at least provide some clues on how autism develops even if it doesn't provide a means of diagnosing it earlier.

  • They didn't predict anything. They retrospectively reviewed scans and determined a "signature" that correlated with the outcome studied. Without an independent validation cohort, this is interesting but far short of definitive. There are concerns about overfitting with such an analysis technique.
  • by Anonymous Coward

    Once autism can be detected in fetuses, they will be aborted routinely.
    Count on it.

    • by quenda (644621)

      Once autism can be detected in fetuses, they will be aborted routinely. Count on it.

      Thankyou Captain Obvious! You saved the day!

  • by Compaqt (1758360) on Thursday February 23, 2012 @10:31PM (#39143447) Homepage

    Assuming autism is mostly genetic (which I don't claim), and assuming they can come up with a prenatal test for it (not this one), would it be moral to abort if a test showed positive for autism?

    • by Prune (557140)
      Of course.
  • because there are study that marrying and being pregnant late can increase the chance of autism in the child.

    There are also there are study that people early marry have higher chance of divorce, bankruptcies, home foreclosure, and delinquent children due to not having the proper financial foundation to begin with.

    I don't know much how autism is being treated here in America, but being acquainted with the eastern culture, tiger-mom tactics and oppressive military-style training are quite popular in combating

    • by Plammox (717738)
      Temple Grandin [wikipedia.org] (american) also advocates the discipline approach. And it makes sense, if no feedback is given to unwanted behaviour, how are the autistic kids else going to learn the social rules invisible to them? It must be applied with sensitivity, though.

The Universe is populated by stable things. -- Richard Dawkins

Working...