Nearly imperceptible fluctuations in movement correspond to autism diagnoses

January 17, 2018, Indiana University
IU Ph.D. student Di Wu directs a volunteer as she touches images on a screen using a device designed to track miniscule fluctuation in the arm's movement. IU-led research suggest physical movement is an accurate method to diagnose neurodevelopmental disorders, including autism. Credit: James Brosher, Indiana University

A new study led by researchers at Indiana University and Rutgers University provides the strongest evidence yet that nearly imperceptible changes in how people move can be used to diagnose neurodevelopmental disorders, including autism.

The study's results, reported Jan. 12 in the Nature journalScientific Reports, suggest a more accurate method to diagnose autism. Current assessments depend on highly subjective criteria, such as a lack of eye or repetitive actions. There is no existing medical test for autism, such as a blood test or genetic screening.

"We've found that every person has their own unique 'movement DNA,'" said senior author Jorge V. José, the James H. Rudy Distinguished Professor of Physics in the IU Bloomington College of Arts and Sciences' Department of Physics. "The use of movement as a 'biomarker' for autism could represent an important leap forward in detection and treatment of the disorder."

It's estimated that 1 percent of the world's population, including 3.5 million children and adults in the United States, are diagnosed with , which is the country's fastest-growing developmental disability.

Unlike diseases diagnosed with medical tests, autism remains dependent upon symptoms whose detection may vary based upon factors such as the person conducting the assessment. The assessments are also difficult to administer to very young children, or to people with impairments such as lack of verbal skills, potentially preventing early interventions for these groups. Early intervention has been shown to play an important role in successful treatment of autism.

"Our work is focused on applying novel data analytics to develop objective neurodevelopmental assessments for autism, as well as other ," said Di Wu, an IU Ph.D. student and the lead author on the study. "We really need to narrow the gap between what physicians observe in patients in the clinic and what we're learning about movement within the field of neuroscience."

To conduct the study, the researchers examined over 70 volunteers as they moved their arm to touch an object on a screen. The volunteers included 30 individuals previously diagnosed with autism, ages 7 to 30, including a girl with no . The group also included 15 neurotypical adults, ages 19 to 31; six neurotypical children; and 20 neurotypical parents of volunteers with autism.

After the assessment, each volunteer was assigned a "score" based on the level of hidden speed fluctuations in their movement. A lower score indicated a greater risk for autism, with numbers under a certain threshold corresponding to previous diagnosis of autism. The greater amount of fluctuation in the movement of the individuals with autism was possibly related to the level of "noise" naturally produced by random neuron firings in the brain, for which neurotypical individuals seem to develop stronger compensation methods.

Eighteen of the 30 individuals in the study with autism were assessed at the IU School of Medicine before the experiment, using four standard psychiatric tests for autism. In each case, the movement-based diagnoses corresponded to these qualitative-based assessments, which are rarely in complete agreement.

The volunteers who scored lower on the scale also exhibited more severe forms of autism. Currently there is no standard accepted quantitative metric to diagnose the disorder's severity. Also, lower-than-average scores in several of the volunteers' parents, who did not have an themselves, suggested that movement could possibly be used to assess a neurotypical parent's risk for children with autism, José said.

The volunteers' movements were captured using high-speed, high-resolution sensors to track fluctuations in movement invisible to the naked eye. The study also tracked changes in speed and position of the arm at every point in movement, as opposed to a single variable—the top movement of the arm's velocity—examined in a previously published study from the team. The new motion data strengthens evidence for movement as a biomarker for autism.

Next, the researchers aim to conduct movement assessments on more people, including more research on the parents of children with autism to better understand the connection between lower parental scores on the movement assessment and their children's risk for .

Explore further: Girls' social camouflage skills may delay or prevent autism diagnosis

More information: Di Wu et al, A Biomarker Characterizing Neurodevelopment with applications in Autism, Scientific Reports (2018). DOI: 10.1038/s41598-017-18902-w

Related Stories

Girls' social camouflage skills may delay or prevent autism diagnosis

January 4, 2018
On parent-reporting measures, girls with autism seem to struggle more than boys with performing routine tasks like getting up and dressed or making small talk, even when the study group is normalized to meet similar basic ...

Time between pregnancies may affect autism risk

November 22, 2017
Investigators have found a link between the amount of time between pregnancies and Autism Spectrum Disorder in children. The findings are published in Autism Research.

Autism severity detected with brain activity test

July 25, 2017
UCLA researchers have discovered that children with autism have a tell-tale difference on brain tests compared with other children. Specifically, the researchers found that the lower a child's peak alpha frequency—a number ...

New iPad game could help diagnose autism in children

August 30, 2016
Autism could be diagnosed by allowing children to play games on smart phones and tablets, a study involving the University of Strathclyde has found.

Autism may be overdiagnosed in the United States

October 27, 2015
(HealthDay)—As many as 9 percent of American children diagnosed with autism may not have the disorder, according to a federal government study published online Oct. 20 in Autism.

Minute movements of autistic children and parents provide clue to severity of disorder

December 1, 2014
Imperceptible variations in movement patterns among individuals with autism spectrum disorder are important indicators of the severity of the disorder in children and adults, according to a report presented at the 2014 Society ...

Recommended for you

Screening may miss signs of autism, especially in girls: study

May 21, 2018
(HealthDay)—An important checklist used to screen for autism can miss subtle clues in some children, delaying their eventual diagnosis.

Autism is not linked to eating fish in pregnacy

May 21, 2018
A major study examining the fish-eating habits of pregnant women has found that they are not linked to autism or autistic traits in their children.

Scientists just beginning to understand autistic adults' unique health needs

May 11, 2018
In the 1990s, the prevalence of autism spectrum disorder (ASD) among children rose sharply. These children are now entering adulthood, yet physicians and scientists know very little about the health outcomes they might face. ...

Meet Nao, the robot that helps treat kids with autism

May 9, 2018
(HealthDay)—It may seem counterintuitive, but a robot might help kids with autism interact better with humans.

New study links strong pupillary light reflex in infancy to later autism diagnosis

May 7, 2018
A new study published in Nature Communications shows that infants who are later diagnosed with autism react more strongly to sudden changes in light. This finding provides support for the view that sensory processing plays ...

Scientists find possible autism biomarker in cerebrospinal fluid

May 2, 2018
Autism diagnosis is slow and cumbersome, but new findings linking a hormone called vasopressin to social behavior in monkeys and autism in people may change that. Low vasopressin in cerebrospinal fluid was related to less ...

1 comment

Adjust slider to filter visible comments by rank

Display comments: newest first

BENRAS
not rated yet Jan 18, 2018
High resolution sensor driven data might be applied to movement characterizing onset of other disease processes ordinarily invisible during onset. Gait and ordinary movement might display markers as well, and speech patterns might also be investigated for characterization,

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.