MU researchers close in on possible autism detection method
Early detection of autism in children is key for treatment, and researchers at the University of Missouri are inching closer to determining the viability of using facial measurements as a potential detection marker.
Ye Duan, associate professor of computer science, worked with postdoctoral fellow Tayo Obafemi-Ajayi, Assistant Professor of Pathology Kristina Aldridge, as well as Judith Miles and T. Nicole Takahashi of MU’s Thompson Center for Autism and Neurodevelopmental Disorders. The team’s work resulted in a paper, “Facial structure analysis separates autism spectrum disorders into meaningful clinical subgroups,” which recently was published in the Journal of Autism and Developmental Disorders.
The paper is a continuation of the 2011 paper “Facial phenotypes in subgroups of pre-pubertal boys with autism spectrum disorders are correlated with clinical phenotypes.” That work used straight-line, or Euclidean, measurements to determine distance between standardized facial landmarks in a group of autistic children and a control group. This time, the geodesic distances were measured via three-dimensional imaging, measuring along the curvature of the face rather than in a straight line.
The data examined in the paper revealed three different clusters of “relatively distinctive clinical and behavioral traits.” One of the three subgroups categorized by this most recent study exhibited similar characteristics to one of the groups in the 2011 study, and this most recent paper posits that since similar results were reached using both Euclidean and geodesic measurements and different cluster techniques, facial measurements may provide a biomarker for autism.
“We want to define the autistic face,” Duan said. “There’s probably not a single autistic face, but we want to try to find how many different main types of the face there are, if there are any.”
The process began back in 2009 with a grant from the Department of Defense’s Defense Medical Research and Development Program. Duan said Miles noticed similarities in facial features among some autistic children and thought perhaps there was something more than coincidence at play.
“Dr. Miles told me she noticed a subgroup of the face of autistic children looks very beautiful and looks very different,” said Duan. “She said maybe there was something we can do on that.”
Strict limitations were placed on participants for inclusion in the study. The children selected were between 8 and 12 years old, male and Caucasian. There were 62 children in the autistic group as diagnosed by the Thompson Center, and 36 non-autistic children in the control group. The goal was to eliminate the variable of different facial features between different ethnic groups and genders, with the hopes that after determining the markers in one group, markers can later be determined for the other groups individually.
It’s a long process, but Duan said he’s hopeful the work inevitably will lead to a viable early detection system for the maximum amount of children possible.
“Potentially, it could be used for early detection screening,” he said. “We need a bigger study, and I hope we have the opportunity to do another round. The next time, we want to have a bigger range, more children and hopefully validate (this study).”
- Computers & Electronics
- Health / Medicine
- Infrastructure & Transportation
- Nano Science & Technology
- National Security / Defense
- The Environment
- All Academic Departments
- Chemical Engineering
- Civil & Environmental Engineering
- Computer Science
- Electrical & Computer Engineering
- Industrial & Manufacturing Systems Engineering
- Information Technology
- Mechanical & Aerospace Engineering
- MU Informatics Institute
- Naval Sciences
- Nuclear Engineering Program
- Nuclear Science & Engineering Institute
- Back to menu
- Faculty & Staff
- Research Centers & Programs
- Mizzou Engineer Magazine
This story is tagged as:
- Transcription app earns students seat at Silicon Valley table
- Catching them before they fall: MU gait study aims to predict fall risk in elderly
- MU researchers monitor health with bed sensors
- Collaborative effort leads to high-resolution imaging breakthrough
- Professor’s algorithm may lessen stress on wireless networks