Sen-Ching Cheung, an associate professor of electrical and computer engineering at the University of Kentucky, never expected to become an autism expert.
But Mr. Cheung, the father of a 5-year-old boy with autism, has seen his career take a twist that mirrors the unpredictable nature of the disease itself: He is putting his digital-imaging skills to work on what he hopes will be a promising technological therapy for autistic children. He is one of a number of scientists seeking federal support for their approaches to autism research, which has an increasingly vocal public constituency and is nearing what could be crucial advances.
In the United States, autism affects an estimated one million children, who have a bewildering array of symptoms and severities. Mr. Cheung’s son, for instance, can be turned frantic by minor acts like having his hair cut or his seat belt buckled. He doesn’t play simple games or even look into his parents’ eyes. “He’s classical autistic,” Mr. Cheung says.
Because of his son, Mr. Cheung grew painfully familiar with a standard therapy known as video self-modeling, in which a child is shown a video of himself performing a daily task—such as brushing his teeth—but only after all the scenes of him being coaxed and prodded have been edited out.
The video self-modeling technique is an approach with “pretty impressive results,” Mr. Cheung says. Still, he says, the cost and complexity of making the videos—one for each child and each task—often proves prohibitive. He wondered if he could use his digital-manipulation skills to create a way of easily importing images of any individual child into videos that would then show that child accomplishing particular tasks.
Mr. Cheung had honed his computer video skills at Kentucky by working on a contract with the Department of Homeland Security involving surveillance technologies. Now he leads an interdisciplinary team at the university that brings together experts in autism, communications sciences, and school psychology, and he is seeking a $726,000 grant from the National Science Foundation to bring his idea to life.
Mounting Diagnoses
The idea could be a breakthrough for wider adoption of the video self-modeling approach, says Kevin M. Ayres, an associate professor of special education at the University of Georgia. Making such videos through digital manipulation, says Mr. Ayres, who also develops computer-based interventions for patients with autism, could be “pretty powerful stuff.”
Federal health officials are considering ideas such as Mr. Cheung’s—and the direction for research that they represent—at what may be a key moment in the fight against autism. Late in March, the federal Centers for Disease Control and Prevention reported finding a 23-percent surge in the number of autism-related disorders from 2006 to 2008, meaning that an estimated one in 88 American children has autism.
Federal financial support for autism research is significant, around $230-million in 2009, but even so, advocates for autism research would like that amount to be higher. “The funding is definitely out of whack,” says Alycia Halladay, director of research for environmental sciences at Autism Speaks, a leading advocacy-and-research group. “There are certainly disorders that affect fewer people that receive more federal funding.”
And within the federal autism-research budget, some scientists note a far heavier commitment to research aimed at diagnoses, causes, and possible cures than to research on helping current patients. “If the funding levels for treatment and intervention improved,” Mr. Ayres says, it would “benefit society at large.”
Of the total federal research spending on autism, the largest portion by far—$197-million in 2009, the latest year for which the government has itemized figures—was distributed by the National Institutes of Health. The CDC got about $19-million. The next-largest shares, each less than $9-million, went to the Health Resources and Services Administration, the Department of Defense, and the Department of Education, where the money largely covered services.
The distribution of money is part of a debate fraught with emotion, as might be expected with a disease that has such devastating impacts on children. Autism is notorious for having a wide range of effects—some people are only mildly impaired, while others spend their lives unable to care for themselves. Congress tried to calm tensions in 2006 by creating the Interagency Autism Coordinating Committee, in which a wide range of parties, including parents and advocacy groups, help set research priorities.
Putting into effect the panel’s entire agenda would cost “a whole lot more” than the amounts approved by Congress, says Catherine E. Rice, an epidemiologist and developmental psychologist in the CDC’s National Center on Birth Defects and Developmental Disabilities. But the existence of the committee is itself a tangible sign of the shared commitment among parties, she says.
Face Time
It’s a world that Mr. Cheung never expected to enter when he arrived at the University of Kentucky, in 2004. One of his first projects was financed by Homeland Security, which wanted to solve a problem involving video surveillance.
The goal was to protect the privacy of workers in job sites that require the extensive use of security cameras. The workers were trusted, but the cameras meant that their daily movements were subject to constant observation as part of the process for identifying possible criminal behaviors of others.
An initial solution involved technologies that digitally blurred the workers’ presence or covered them with a black box. Neither was optimal, Mr. Cheung said. Blurring, for instance, hindered computerized analyses of the scenes. And both methods left enough clues to a worker’s identity that the surveillance still had the effect of monitoring his or her minute-by-minute activity.
The solution developed by Mr. Cheung and his team completely erased designated people from the videos. The technology uses a statistical analysis of the background to generate a scene without any sign at all of the trusted worker.
About three years ago, Mr. Cheung and his wife began to have concerns about their son, then two years old. The child would avoid looking at him, the professor says, even during close interactions such as bathing. He also wouldn’t participate in simple games like hide-and-seek. “He was just not responding,” Mr. Cheung says.
A series of doctor visits, first with pediatricians and then with autism experts, eventually confirmed the diagnosis. The process led Mr. Cheung to a colleague, Neelkamal S. Soares, an assistant professor of pediatrics at Kentucky who suggested various behavioral techniques that other parents had used successfully, including video therapies.
A simple version involves showing children videos of child actors performing daily tasks. Mr. Cheung’s son, however, would not even watch the videos. That’s common, Mr. Cheung says, because children with autism often can’t imagine themselves in the position of others. Another problem for his ethnic Chinese son, he says, may have been that the children in the videos, many with blond hair and blue eyes, “don’t look anything like him.”
Dr. Soares, a native of India, knew of no such videos produced in Asia. “We got thinking,” Mr. Cheung says. “Since I do computer graphics, could we do this like we do in Hollywood, but with my son doing these proper behaviors, so he can watch himself?”
Technological Aids
Ms. Rice knows of other instances of parents of children with autism using professional skills to try to help. They include Gregory D. Abowd, a professor of interactive computing at the Georgia Institute of Technology, who is director of the Health Systems Institute there. He created the Aware Home Research Initiative at Georgia Tech, an interdisciplinary center for solving the technological and social challenges facing people who need help living independently in a home setting.
There’s also a burgeoning interest in finding ways to use technology to help people with autism, One possibility is creating video systems that could watch a room full of children to identify those with telltale signs. Another is developing smartphone apps that could guide people through daily activities with which they have difficulty, such as riding a bus or heating a frozen dinner.
“What he’s proposing is very interesting,” Ms. Rice says of Mr. Cheung’s idea, which he is developing with Lisa A. Ruble, an associate professor of school psychology at Kentucky; Ramesh S. Bhatt, a professor of psychology; and Rita R. Patel, an assistant professor of communication sciences and disorders.
Still, Ms. Rice and others caution, there’s a long road to proving that an approach is effective. And still others, such as Nancy J. Minshew, a professor of psychiatry and neurology at the University of Pittsburgh, are wary of raising hopes too fast or too early. With increasing attention to autism, there’s an ever-growing industry trying to sell cures to families, sometimes at high cost with no scientific benefit, Dr. Minshew says. “Families are besieged with these great things that they want to go out and buy tomorrow that may or may not work.”
Time will tell if Mr. Cheung has a real breakthrough. “He’s a parent,” Dr. Minshew says, “and what he tailor-makes to his own child is only one view of autism.
“There’s a saying in autism that if you’ve met one person with autism, you’ve met one person with autism,” she says. “Everybody’s needs can be so different.”
