Rapid advancements in artificial intelligence technologies offer new possibilities for early screening, diagnosis, and intervention processes in Autism Spectrum Disorder. Asst. Prof. Demet Gülaldı from the Child Development Department stated that artificial intelligence systems can analyze patterns in children's eye movements, voice tone, facial expressions, attention processes, and play behaviors, adding, “Artificial intelligence is no longer just a technology; it is becoming an important tool for early diagnosis, early intervention, and receiving proper support in the field of autism.”
Emphasizing the importance of early intervention in autism, Dr. Gülaldı said, “Artificial intelligence-supported systems can analyze children's behaviors, evaluate eye contact, facial expressions, and attention processes, and thus significantly shorten the diagnosis period.”
Asst. Prof. Demet Gülaldı from the Child Development Department of Üsküdar University's Faculty of Health Sciences evaluated the developments at the intersection of artificial intelligence and Autism Spectrum Disorder, and the opportunities artificial intelligence offers in early screening, diagnosis, and intervention processes for autism.
Artificial intelligence now also analyzes patterns in human behavior
Asst. Prof. Demet Gülaldı pointed out that the intersection of artificial intelligence and Autism Spectrum Disorder is one of the rapidly developing fields of study in recent years, stating, “It is no longer possible to see artificial intelligence (AI) merely as a data processing system. Today, these systems can learn, analyze human behaviors, and solve highly complex patterns in a short time. Especially developments in Deep Learning and Natural Language Processing (NLP) enable machines not only to execute commands but also to understand the nuances in human behavior. Today, AI can make diagnoses at a level comparable to medical diagnoses, even catching microscopic clues within massive data heaps that the human eye might miss.”
Diagnosis in autism largely relies on behavioral indicators
Asst. Prof. Demet Gülaldı stated that autism is a neurodevelopmental condition characterized by social interaction, communication, and behavioral patterns, spanning a wide spectrum. She added, “Autism poses a serious threat to individuals' physical and mental health, leading to significant impairments in social interaction, learning, work life, and family life. The causes of autism are complex and multifaceted, including genetic and environmental factors. Although the disorder has some genetic foundations, clinical diagnosis today largely relies on observable and measurable behavioral indicators due to the high individual differences in molecular genetic indicators, such as susceptibility genes.”
Early intervention is critically important for development
Highlighting the importance of early diagnosis and intervention in autism, Asst. Prof. Demet Gülaldı said, “Early diagnosis intervention is effective in reducing deficiencies in social attention, language, and intelligence in individuals with autism, and in decreasing the overall severity of symptoms. Although autism symptoms typically appear before the age of 2, a significant portion of individuals with autism are diagnosed after the age of 6 for various reasons, which leads to missing the optimal period for behavioral therapy and negatively impacts cognitive development and quality of life.”
Numerous data, from eye movements to voice tone, can be analyzed
Asst. Prof. Demet Gülaldı stated that the combination of artificial intelligence and autism research reveals significant opportunities for early screening, remarking, “When artificial intelligence and autism studies come together, we see that technology and technological advancements become very important in the lives of individuals with autism. Machine learning algorithms being able to detect signs that traditional methods catch at age 6, from a child's eye movements, frequency changes in voice tone, or play style, at 6-12 months of age are very significant developments.”
Asst. Prof. Demet Gülaldı also expressed that artificial intelligence-supported intervention tools could create the opportunity to provide a personalized and patient learning environment suited to each child's unique "operating system." She added, “This combination will make it easier for us to position autism not as a disability, but as a valuable difference that can be integrated into society with the right technological support. In light of scientific research conducted in recent years, we can say that artificial intelligence has great developmental potential in the early screening, diagnosis, and early intervention processes of autism in young children and surpasses traditional methods.”
Artificial intelligence can speed up the diagnosis process
Dr. Gülaldı reminded that information obtained from parents and clinical observations play an important role in traditional early screening processes, and continued:
“Traditional methods used for early screening of children with autism involve psychiatrists identifying autism symptoms in a child's development based on parent reports and clinical observations. However, artificial intelligence methods such as machine learning can automate and accelerate the autism diagnosis process by learning data categories from numerous past autistic samples and control groups. For example, promising results are reported for Convolutional Neural Network (CNN)-based ASD classification systems in terms of performance metrics such as accuracy, sensitivity, and specificity. These methods enable the creation of predictive models that can classify new input samples.”
Studies focus on its use especially in young children
Dr. Gülaldı stated that there are studies on the use of AI-based models, especially in young children, and said, “A study has shown that by creating an alternating decision tree model with artificial intelligence, it can effectively diagnose autism in children aged 13 to 48 months with a 99.97 percent accuracy rate. It has been proven that artificial intelligence can help young children with autism engage in basic social interactions and interpersonal communication, and develop their abilities to recognize and manage emotions.”
Personalized support for every child may be possible
Dr. Gülaldı stated that artificial intelligence can provide significant opportunities not only in early screening and diagnosis processes but also in intervention and education. She concluded, “Artificial intelligence-supported systems can analyze children's behaviors, evaluate eye contact, facial expressions, and attention processes, and thus significantly shorten the diagnosis period. Therefore, it is encouraging that these systems offer critical opportunities for early intervention in autism. Artificial intelligence is no longer just a technology; it is becoming an important tool for early diagnosis, early intervention, and receiving proper support in the field of autism.”
