Indianapolis, Indiana – In Indiana, where it can take up to 18 months for children with suspected autism spectrum disorder (ASD) or attention-deficit hyperactivity disorder (ADHD) to secure a diagnostic appointment with a psychiatrist, a breakthrough in artificial intelligence (AI) is offering hope for faster and more accurate diagnoses. Researchers at Indiana University (IU) have developed a promising new diagnostic approach that leverages AI to speed up the identification of neurodivergent disorders, such as autism and ADHD, while improving the accuracy of those diagnoses. The development could not only reduce wait times but also help streamline the diagnostic process, allowing for quicker intervention and treatment for children who need it.
Currently, psychiatrists rely on a variety of subjective tests and patient surveys to diagnose neurodivergent disorders, looking at symptoms like communication difficulties, hyperactivity, and repetitive behaviors. However, as experts point out, there is no widely available quantitative or biological test that can definitively diagnose conditions like autism or ADHD. As Dr. Jorge José, the James H. Rudy Distinguished Professor of Physics at IU Bloomington, explained, “The symptoms of neurodivergent disorders are very heterogeneous; psychiatrists call them ‘spectrum disorders’ because there’s no one observable thing that tells them if a person is neurotypical or not.” These challenges have long hindered the speed and reliability of diagnoses.
With that in mind, Dr. José, alongside an interdisciplinary team of scholars from IU, including Distinguished Professor Emeritus John I. Nurnberger and Associate Professor of Psychiatry Martin Plawecki, set out to address this gap in diagnosis. Their research, published on July 8 in Nature’s Scientific Reports, details their groundbreaking work in using AI to speed up and enhance the accuracy of diagnosing autism and ADHD.
Read also: UIndy biology students explore spider diversity during 24-hour BioBlitz at Eagle Marsh
The team’s approach utilizes artificial intelligence to analyze movement biomarkers, or subtle movements that are often imperceptible to the naked eye but measurable using advanced sensors. This AI-driven method could potentially diagnose autism or ADHD in as little as 15 minutes. Khoshrav Doctor, a Ph.D. student at the University of Massachusetts Amherst and a former research scholar at IU, explained, “It could help as an additional tool in the clinician’s toolbelt. It also gives us the ability to see who might need the quickest intervention and direct them to providers earlier.”
The research team’s journey into using AI for diagnosis began in 2018 when Dr. José’s collaboration with Rutgers led to the discovery of movement biomarkers associated with autism. In a groundbreaking study, the researchers used sensors to capture participants’ hand movements as they reached for a target on a computer screen. The data revealed that neurotypical individuals moved in a distinctly different way compared to those with autism. Specifically, participants with autism displayed increased randomness in their movements, a finding that suggested the potential of using movement patterns as a diagnostic tool.
Since then, the team has improved their diagnostic methodology by using high-definition kinematic Bluetooth sensors, which allow for the collection of more detailed movement data. Rather than just measuring the velocity of movements, the new sensors track acceleration, rotation, and other movement variables. Chaundy McKeever, a physics graduate student at IU, explained, “We’re taking a physicist’s approach to looking at the brain and analyzing movement specifically.” Through this refined technique, the team has observed that more sporadic or erratic movements often correlate with more severe cases of neurodivergent disorders.
One of the most significant advancements in their recent research is the integration of deep learning, a specialized area of artificial intelligence, into the diagnostic process. By using supervised deep learning, the team has been able to analyze raw movement data from participants with ASD, ADHD, a combination of both, and neurotypical development. This enhanced AI method allows researchers to better assess the severity of a neurodivergent disorder, which could prove invaluable in treatment planning and monitoring.
Dr. José elaborated on the significance of this breakthrough, saying, “By studying the statistics of the motion fluctuations, invisible to the naked eye, we can assess the severity of a disorder in terms of a new set of biometrics. No psychiatrist can currently tell you how serious a condition is.” With the ability to more accurately assess the severity of a disorder, healthcare providers can tailor treatments more effectively. For instance, patients with more severe symptoms may require specialized care, while those in the middle of the spectrum could receive less demanding treatments, which could often be administered at home, reducing the cost and complexity of care.
Moreover, the AI method developed by Dr. José and his team could be used in school settings to quickly identify children who may need further evaluation. This would allow educators and healthcare providers to intervene much sooner, preventing delays that could negatively affect the child’s development. The team envisions a system in which schools can use AI to screen students for neurodivergent disorders, triaging those who need more in-depth evaluations or treatment. According to Doctor, “It could be used in schools to triage students who might need further care.”
However, it’s important to note that the AI approach is not intended to replace psychiatrists in the diagnostic process. Instead, it serves as an additional tool to enhance the overall diagnostic process, making it faster and more accurate. This collaborative approach, which combines AI with the expertise of trained medical professionals, aims to improve the way neurodivergent disorders are diagnosed and treated. “It could help as an additional tool in the clinician’s toolbelt,” said Doctor, emphasizing the supportive role AI plays in the process rather than supplanting traditional methods.
The potential impact of this research is far-reaching, offering hope not only to families struggling with lengthy wait times for diagnoses but also to healthcare providers looking for ways to improve the precision of their diagnoses. By combining advanced AI with motion biometrics, Dr. José and his team are paving the way for a future where diagnosing autism, ADHD, and other neurodivergent disorders is quicker, more reliable, and more effective.
As the study continues to gain attention, it is clear that this new approach could revolutionize the way we think about diagnosing and treating neurodivergent disorders. The ability to quickly assess and intervene in a child’s development can lead to better outcomes and improved quality of life for children with autism and ADHD. With the continued advancement of AI and its application to healthcare, the future of neurodivergent disorder diagnoses looks promising—one where speed, accuracy, and accessibility are no longer limited by traditional diagnostic methods.
