Researchers use artificial intelligence to

Image: Han-Seok Seo
see After

Credit: University Relations

Could artificial intelligence be used to help early detection of autism spectrum disorders? That’s a question researchers at the University of Arkansas are trying to answer. But they take an unusual approach.

Han-Seok Seo, Associate Professor with a Joint Appointment in Food Science and UA System Agriculture Division, and Khoa Luu, Assistant Professor in Computer Science and Computer Engineering, will identify sensory cues of various foods in neurotypical children and those known to be on the spectrum. Machine learning technology will then be used to analyze biometric data and behavioral responses to these smells and tastes to detect indicators of autism.

There are a number of behaviors associated with ASDs, including difficulties with communication, social interaction, or repetitive behaviors. People with ASD are also known to exhibit abnormal eating behaviors, such as avoidance of some – if not many – foods, specific meal requirements, and unsocial eating. Food avoidance is of particular concern, as it can lead to poor nutrition, including vitamin and mineral deficiencies. With this in mind, the duo intends to identify sensory cues from food that trigger atypical perceptions or behaviors upon ingestion. For example, smells like peppermint, lemons, and cloves are known to evoke stronger reactions in people with ASD than others, possibly triggering heightened levels of anger, surprise, or disgust.

Seo is an expert in the fields of sensory science, behavioral neuroscience, biometric data, and eating behavior. He organizes and directs this project, including screening and identifying specific sensory cues that can differentiate autistic children from non-autistic children with respect to perception and behavior. Luu is an artificial intelligence expert specializing in biometric signal processing, machine learning, deep learning, and computer vision. He will develop machine learning algorithms to detect ASD in children based on unique patterns of perception and behavior in response to specific test samples.

The duo is in the second year of a three-year, $150,000 grant from the Arkansas Biosciences Institute.

Their ultimate goal is to create an algorithm that exhibits equal or better performance in early detection of autism in children compared to traditional diagnostic methods, which require trained health and psychology professionals to perform assessments. , longer assessment times, caregiver-submitted questionnaires, and additional medical examinations. costs. Ideally, they will be able to validate a lower-cost mechanism to aid in the diagnosis of autism. While their system is unlikely to be the final word in a diagnosis, it could provide parents with an initial screening tool, ideally weeding out children who are not candidates for ASD while ensuring that the most likely candidates go through a process of more comprehensive screening.

Seo said he became interested in the possibility of using multisensory processing to assess ASDs when two things happened: He started working with a graduate student, Asmita Singh, who had experience in work with autistic students, and the birth of his daughter. Like many first-time parents, Seo paid close attention to her newborn, anxious for her to be healthy. When he noticed she wouldn’t make eye contact, he did what most nervous parents do: turned to the internet for an explanation. He learned that avoiding eye contact was a known characteristic of ASDs.

Although her child did not end up having ASD, her curiosity was piqued, particularly about the role smell and taste sensitivities play in ASD. Further conversations with Singh led him to believe that other anxious parents could benefit from an early detection tool – perhaps to ease worries at little cost upfront. Subsequent conversations with Luu led the pair to believe that if machine learning, developed by his graduate student Xuan-Bac Nguyen, could be used to identify normal reactions to food, it could also be taught to recognize atypical responses. .

Seo is looking for volunteers between the ages of 5 and 14 to participate in the study. Neurotypical children and children already diagnosed with ASD are required for the study. Participants receive a $150 electronic gift card for their participation and are encouraged to contact Seo at [email protected]

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of press releases posted on EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.

Comments are closed.