Researchers are currently exploring the idea that artificial intelligence (AI) could be used to diagnose and treat autism spectrum disorder (ASD). An ASD diagnosis can be tricky, and for some it can take years. Now a new app developed at Duke University may be able to spot the disorder in 10 minutes while children play on a tablet, while other AI apps are helping children who are already diagnosed.
Duke University Doctors Develop an App to Help Screen for Autism
Researchers at Duke University have developed an AI-driven app that can accurately detect a range of behaviors associated with ASD. To use the app, called Sense to Know (S2K), children play a bubble-popping game. The app measures and weighs a variety of behavior indicators for ASD, accurately reflecting the complexity in each child. The app delivers scores that evaluate the probability that the child tested is on the autism spectrum and which of the behavioral indicators led to its conclusions.
To diagnose ASD, S2K detects a wide range of behaviors such as facial expression, gaze patterns, head movements, and blink rate. The on-screen bubble-popping game assesses motor movement and skills, as delays in motor skills are one of the earliest signs of ASD. The app uses almost every sensor in the tablet’s arsenal to measure and characterize the child’s response without the need for any sort of special equipment. AI can use the data to analyze how likely it is that a child will be diagnosed with ASD.
The effectiveness of the app was demonstrated in a study published online this month in the journal Nature Medicine.
How the App Was Tested
To study the effectiveness of the app, 475 children, ages 17–36 months, were evaluated during pediatric well-child visits. Of that group, 49 were later diagnosed with ASD and 98 with a developmental delay without ASD. The app showed 87.8 percent sensitivity for detecting autism, meaning it correctly identified most children with the condition. The percentage of children without autism who screened negative was 80.8 percent.
Overall, participants who screened positive for ASD using the app had a 40.6 percent probability of being subsequently diagnosed with the condition. Although this may seem like a low percentage, currently, only about 15 percent of children who screen positive with standard parent questionnaires are later diagnosed with ASD. Combining the app with the standard questionnaire increased the percentage of positive screenings followed by a later diagnosis to 63.4 percent!
“The AI we’ve built compares each child’s biomarkers to how indicative they are of autism at a population level,” said Sam Perochon, a Ph.D. student at Duke working on the study. “This allows the tool to capture behaviors other screening tests might miss and also report on which biomarkers were of the most interest and most predictive for that particular child.” In a university news release, co-lead researcher Geraldine Dawson said “autism is characterized by many different behaviors, and not all children on the spectrum display all of them equally, or at all. This screening tool captures a wide range of behaviors that more accurately reflect the complexity and variability found in autism.”
“Just like when any patient goes to their doctor, the doctor listens to them describe what they are experiencing, but they also use thermometers and other objective tests to provide additional information to guide next steps and referrals for further evaluation,” added Dawson. “Such objective tests have been missing for autism.”
The researchers are currently conducting a study in which parents deliver the app at home. They hope that the app will also be useful for measuring a child’s progress within an early intervention program as well as for studying the effectiveness of such programs.
“There is a wide range of expertise among health care providers in knowing and being able to recognize all the potential signs of a child being on the autism spectrum,” Dawson said. “This app could help clinicians focus on the areas in which the child needs help, as well as identify areas of strength.”
Other AI Apps for Diagnosis and Treatment of ASD
There are a few companies who are also pioneering methods to diagnose autistic children using AI and AI-like technology.
Behavior Imaging developed a system called the Naturalistic Observation Diagnostic Assessment. This tool is an app that allows parents to upload videos of their children for observation. Clinicians watch the videos to make remote diagnoses. More recently, the company has started training AI-like algorithms to observe and categorize behaviors. The algorithms would not diagnose the children, but they might be used to point clinicians to specific behaviors that might otherwise have been missed.
Another use of AI-aided diagnosis is an autism screening tool created by Cognoa, which was approved by the US Food and Drug Administration. This tool is a mobile app that parents can use without the involvement of a trained evaluator. It reviews answers to multiple-choice questions, as well as videos of the child.
Treatment Apps
Besides the diagnostic apps, there are many autism apps on the market that support behavioral therapy and learning. Here are some examples:
The Manatee app is one of the first AI apps offered as a simple, no-cost iPhone download. Manatee uses a series of exercises to help the user calm down or respond appropriately and then, depending on the mood of the child, the model offers exercises and then learns how the child responds.
An AI-powered app created by SocialMind Autism gives parents and caregivers practical guidance on moments of attempted verbalization they may have otherwise overlooked. The app mainly targets children under five, as those years are the most critical in building language skills.
BeMe.ai combines wearable technology and a smartphone app to allow parents, caregivers, therapists, and physicians to gain new insights into a child’s health, behavior, and overall progress.
The Importance of Early Diagnosis of ASD and Intervention
There is no cure for ASD, and there’s no consensus on what causes it. However, there is a lot of evidence that early diagnosis and intervention can improve social and communication skills during infancy, and this has the potential to significantly help improve the child’s later development.
Proper early treatment can reduce children’s symptoms and can improve their overall development by helping them learn new skills that will allow them to be more independent throughout their life. Because they are receiving appropriate treatment at key developmental stages, children with ASD are more apt to gain essential social skills and the ability to act more naturally in social situations. These are valuable tools that will help the children to be more independent and require fewer services as they age.
Early diagnosis can also benefit parents. Instead of noticing symptoms and worrying that something is wrong with their child, an early diagnosis enables parents to take action and begin helping their child and planning for the future. Hopefully, the AI diagnostic and treatment apps described will be helpful in early diagnosis and intervention.
Planning in Advance for Special Needs Children
A special needs trust is an essential tool to protect the financial future of an individual with special needs. Also known as a supplemental needs trust, this type of trust preserves eligibility for federal and state benefits by keeping assets out of the name of the person with special needs, while still allowing the trust funds to be used to benefit the person with special needs.
Planning for a child with special needs is about your child getting the best possible care after your death or when you become unable to make decisions for your child because of your own incapacity or because your child turns 18. For children with special needs who have turned 18 and are capable of signing incapacity planning documents such as a power of attorney and an advance medical directive, it is strongly suggested that they sign these documents as soon as possible after they turn 18. Although children with severe autism may require guardianship and conservatorship, this should generally be a last resort.
When it comes to special needs planning, estate planning, and retirement planning, the attorneys at the Farr Law Firm can guide you through this process.
Special Needs Attorney Fairfax: 703-691-1888
Special Needs Attorney Fredericksburg: 540-479-1435
Special Needs Attorney Rockville: 301-519-8041
Special Needs Attorney DC: 202-587-2797
About Evan H Farr, CELA, CAP
