2024-08-20 03:39:52
A new machine learning model can predict autism in young children from limited information. This is shown by a new study Instituto Karolinska (Sweden) published in ‘JAMA Network Open‘. The model can facilitate the early detection of autism, which is important for providing appropriate support.
“With an accuracy of almost 80% for children under the age of two, we expect it to be a valuable tool for health,” said Kristiina Tammimies, author of the study.
The team used a large US database (SPARK) with information on nearly 30,000 people with and without autism spectrum disorders.
By analyzing a total of 28 different parameters, they developed four different machine learning models to identify patterns in the data. The selected parameters are information about children that can be obtained without extensive evaluations or medical tests before 24 months of age. The model with the best performance is called «AutMedAI«.
In a group of nearly 12,000 people, the AutMedAI model was able to identify approximately 80% of children with autism. In specific combinations with other parameters, age at first smile, first short sentence, and presence of eating problems are strong predictors of autism.
«The results of the study are important because they show that it is possible to identify individuals who may have autism from limited and readily available information.», explained the first author of the study Shyam Rajagopalan.
Basic analysis
Early detection is important, according to the researchers, to make effective interventions that can help children with autism in the development deficit.
“This can dramatically change the conditions for early diagnosis and interventions and ultimately improve the quality of life for many people and their families,” said Shyam Rajagopalan.
In the study, the AI model showed good results identifying children with more complex problems in social communication and cognitive ability and with more general delays in development.
The research team is now planning to make further improvements and validate the model in clinical settings. Work is also underway to incorporate genetic information into the model, which may result in even more specific and accurate predictions.
#artificial #intelligence #predicts #risk #autism