Will artificial intelligence soon play a role in diagnosing autism spectrum disorders?
For children with autism and their families, early diagnosis and intervention can play a big role in improving development and long-term health. Early diagnosis usually leads to earlier access to clinical, behavioral, educational, and social services that a child and family may need. But many families – especially in rural, low-income and minority backgrounds – face long delays in receiving diagnoses and services. These delays can greatly affect a child’s behavior and brain development with lifelong consequences.
A quarter of children under the age of eight living with autism are not currently diagnosed, according to a recent study by Rutgers University professor Walter Zagorodny, and most are black or Hispanic. Meanwhile, although autism is more common in girls than previously thought, boys are still four times more likely to be diagnosed than girls, and boys are also diagnosed earlier than girls. For race and ethnicity, socioeconomic status, geography and gender, the story is the same: children from less affluent and minority groups are diagnosed and start individual treatment later than Caucasian, male and wealthier children.
Despite these realities, I am optimistic that the healthcare community is increasingly aware of the existing imbalances and is ready to adopt technologies designed to address these challenges so that we can improve the lives of children and families living with autism.
Behind the disparities
As Sarabet Broder-Fingert, a pediatrician at Boston University, points out, most autism research focuses on “white children and higher-income families,” and primarily on young white men. This means that when children, especially non-white men, are assessed for autism, the assessments are largely based on data that is not always representative of all children. This has real implications for non-white children and their families.
Exacerbate disparities in access to health resources, different levels of education and a broad understanding of autism, language barriers and more. Understanding the complex health care system is difficult for families, especially those with limited resources.
In an ideal world, pediatricians, who are usually the doctors most often visited by children and families, would have more tools and resources to respond to autism and / or child developmental progress to coordinate appropriate care in a timely manner. effective way.
Click here to learn more
Can AI technology help make this a reality?
Much has been said about the potential of artificial intelligence (AI) in a number of industries. At its core, AI is a way to understand and comprehend vast amounts of information.
In this case, it includes different types of data – videos of children playing, answers to questions from parents and health professionals, clinical data of children with and without autism – which reflect many characteristics of the autism spectrum, such as eye contact and child response to social signals and emotional exchanges.
However, AI does more than just process data. Its true value lies in its ability to identify subtle relationships between different data points. By simultaneously analyzing hundreds or even thousands of data points, AI algorithms can identify and predict behavioral patterns that point to autism or from it.
This helps physicians make more sound and effective assessments, as opposed to a scenario where they rely solely on their own training, experience, and clinical observation of the child. Such observations may be influenced by the child’s gender, race, ethnicity, economic status and geographical location.
With the help of artificial intelligence-based diagnostics, doctors may be able to make more accurate diagnoses. As technology becomes more sophisticated, it can also be used to assess a child’s speech abilities and differences in development.
Using artificial intelligence solutions, we can imagine that diagnosticians will be able to see children with more complex manifestations of autism earlier because those children with clearer diagnoses of autism can be diagnosed and maintained faster by pediatricians.
Technology can help create a more streamlined care system. This will benefit children and families who are more likely to have access to appropriate care. Another consequence could be a more responsive medical community capable of better serving the needs of patients and their caregivers.
Of course, artificial intelligence systems are as good as the data on which they are built. This is why algorithms should be designed to deliberately take into account the gender, racial, ethnic and socio-economic dynamics of the child, recognizing, for example, that girls with autism exhibit different traits than boys with autism.
Including equally subtle data, AI can scale this knowledge and experience in a way that reflects the very diverse communities of children and families, and makes them accessible to every pediatrician. In this way, children and families will finally have more equal access to more accurate and effective diagnoses.
Collaboration and education are key
Artificial intelligence-based diagnosis should become an increasingly important tool in physicians ’toolkits. But technology in autism is not a silver bullet. New solutions in clinical practice should be accompanied by education and training that better informs and connects families and clinicians with networks and services that provide ongoing care and improve outcomes throughout life.
Fortunately, the healthcare community is increasingly attuned to many factors that affect the health of children, families, and society. Along with technological advances and an overall focus on better outcomes, we can ensure that every child and family living with autism receives optimal, timely and sensitive help and support – regardless of gender, family background, location or income.
This article was submitted to Issue 125 – Deployment of ABA Therapy