AI model helps early detection of autism

A new machine learning model can predict autism in young children from relatively limited information, according to a new study from Karolinska Institutet published in Opening of the JAMA networkThe model can facilitate early detection of autism, which is important for providing appropriate support.

“With an accuracy of almost 80% for children under two years old, we hope this will be a valuable tool for health care,” says Kristiina Tammimies, associate professor at KIND, the Department of Women’s and Children’s Health, Karolinska Institutet and last author of the study.

The research team used a large American database (SPARK) containing information on approximately 30,000 people with and without autism spectrum disorders.

By analyzing a combination of 28 different parameters, the researchers developed four separate machine learning models to identify patterns in the data. The selected parameters were information about children that can be obtained without extensive assessments or medical testing before the age of 24 months. The best-performing model was named “AutMedAI.”

Among about 12,000 individuals, the AutMedAI model was able to identify about 80% of autistic children. In specific combination with other parameters, age of first smile, first short sentence, and the presence of eating difficulties were strong predictors of autism.

“The study’s results are significant because they show that it is possible to identify individuals who are likely to have autism from relatively limited and readily available information,” says the study’s first author, Shyam Rajagopalan, a researcher affiliated with the same department at Karolinska Institutet and currently an assistant professor at the Institute of Bioinformatics and Applied Technology, India.

Early diagnosis is essential to implement effective interventions that can help autistic children develop optimally, researchers say.

“This can radically change the conditions for early diagnosis and intervention and ultimately improve the quality of life for many individuals and their families,” says Rajagopalan.

In the study, the AI ​​model performed well in identifying children with greater difficulties in social communication and cognitive abilities and with more general developmental delays.

The research team now plans further improvements and validations of the model in clinical settings. Work is also underway to include genetic information in the model, which could lead to even more specific and accurate predictions.

“To ensure that the model is robust enough to be implemented in clinical settings, rigorous work and careful validation are required. I want to emphasize that our goal is to make the model a valuable tool for health care, and that it is not intended to replace a clinical assessment of autism,” Tammimies says.