A new study from Karolinska Institutet reveals that a new machine learning model can predict autism in young children based on limited information. Early detection is crucial for providing timely support, making this model a potentially valuable tool for facilitating early intervention.
“Kristiina Tammimies, Associate Professor at KIND, the Department of Women’s and Children’s Health at Karolinska Institutet and the last author of the study, states, “With an accuracy of almost 80 percent for children under the age of two, we hope that this will be a valuable tool for healthcare.”
The research team utilized the SPARK database, which contains data on around 30,000 individuals with and without autism spectrum disorders in the US.
The researchers created four different machine-learning models by analyzing 28 different parameters to identify patterns in the data. These parameters included information about children that could be obtained without extensive assessments and medical tests before they reached 24 months of age. The most successful model was called ‘AutMedAI’.
In a study involving around 12,000 individuals, the AutMedAI model successfully identified approximately 80% of children with autism. The age at which a child first smiled, uttered a short sentence, and experienced eating difficulties, when considered in specific combinations with other parameters, emerged as strong predictors of autism.
“The results of the study 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 study first author Shyam Rajagopalan, an affiliated researcher at the same department at Karolinska Institutet and currently an assistant professor at the Institute of Bioinformatics and Applied Technology in India.
Early diagnosis is crucial, according to researchers, for implementing effective interventions to help children with autism develop optimally.
In the study, the AI model demonstrated good results in identifying children with more extensive difficulties in social communication and cognitive ability, as well as having more general developmental delays.
The research team is currently planning additional enhancements and validation of the model in clinical settings. They are also working on incorporating genetic information into the model, which could result in even more specific and accurate predictions.
“To ensure that the model is reliable enough to be implemented in clinical contexts, rigorous work and careful validation are required. I want to emphasize that our goal is for the model to become a valuable tool for health care, and it is not intended to replace a clinical assessment of autism,” says Kristiina Tammimies.
