United States: According to the latest findings, a novel machine learning model called AutMedAI has been formed, which could predict autism in children of less than two, with accuracy almost equal to 80 percent, thus providing an effective tool for early detection.
More about the finding
The study examined twenty-eight parameters made available before twenty-four months, like the age of first smile and eating difficulties, to predict autism.
According to scientists, diagnosing autism is an important task that would impact optimal development. However, further strengthening of the model is currently underway.
About the new machine learning model
The newly formed machine learning model has been able to predict autism in young children on the basis of very limited input about the children under observation.
The new study from Karolinska Institutet has published its findings in JAMA Network Open.
According to Kristiina Tammimies, Associate Professor at KIND, the Department of Women’s and Children’s Health, Karolinska Institutet, who is also an author of the study, “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,” as neurosciencenews.com reported.
In order to perform the study, the research group utilized a large US database (SPARK), having information on almost 30,000 people with and without autism spectrum disorders.
The scientists then studied a combination of 28 different parameters, where they formed four distinct machine-learning models to see patterns in the data.
The selected parameters were information regarding children, which would be fetched without many assessments and medical tests before 24 months of age.
The Model name is ‘AutMedAI,” with the help of which 80 percent of children with autism could be able to be diagnosed with about 12,000 individuals.
As the experts explain, the specific combinations where parameters such as the age of first smile, first short sentence, and the presence of eating difficulties were among the strongest indicators/predictors of autism.
According to Shyam Rajagopalan, the study’s first author and an affiliated researcher at the same department at Karolinska Institutet and currently assistant professor at the Institute of Bioinformatics and Applied Technology, India, “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,” as neurosciencenews.com reported.
Moreover, as the experts mention, early diagnosis plays a vital role in pursuing an effective intervention that would help children with autism develop optimally.
According to Shyam Rajagopalan, “This can drastically change the conditions for early diagnosis and interventions and ultimately improve the quality of life for many individuals and their families.”