Machine Learning Identifies Blood Biomarkers Of Autism by Karishma Abhishek on February 26, 2021 at 10:29 AM Specific group of biomarkers identified in the blood - nine serum proteins using machine learning tools strongly predicts ASD. This may lead to an earlier diagnosis of children with autism spectrum disorder (ASD) that in turn, would help develop more effective therapies, as per research at UT Southwestern, published in PLOS ONE. neuro-developmental delay. It is primarily characterized by social, communication and behavioral challenges. Around 1 in 59 children are diagnosed with autism in the United States, with an average age of 4 years. Earlier diagnosis before the age of 4, trailed by prompt therapeutic support and intervention, could even have a significant effect on their symptoms like inflexible behaviors and the lack of communication or social skills.