AI, Precision Medicine Tool May Enable Early Autism Diagnosis

A precision medicine method enabled by artificial intelligence could lead to the first biomedical screening tool for a subtype of autism, according to a study published in Nature Medicine.

Autism affects an estimated one in 54 children in the US, researchers stated. Boys are four times more likely than girls to be diagnosed. While most children are diagnosed after age four, autism can be reliably diagnosed based on symptoms as early as age two.

Researchers studied a subtype of autism known as dyslipidemia-associated autism, which represents 6.55 percent of all diagnosed autism spectrum disorders in the US.

"Previously, autism subtypes have been defined based on symptoms only -- autistic disorder, Asperger syndrome, etc. -- and they can be hard to differentiate as it is really a spectrum of symptoms," said study co-first author Dr. Yuan Luo, associate professor of preventive medicine: health and biomedical informatics at the Northwestern University Feinberg School of Medicine.

"The autism subtype characterized by abnormal levels identified in this study is the first multidimensional evidenced-based subtype that has distinct molecular features and an underlying cause."

To get an accurate representation of the real world, the researchers layered various data elements on top of one another.

"Our study is the first precision medicine approach to overlay an array of research and health care data -- including genetic mutation data, sexually different gene expression patterns, animal model data, electronic health record data and health insurance claims data --and then use an AI-enhanced precision medicine approach to attempt to define one of the world's most complex inheritable disorders," said Luo.

Researchers then identified clusters of gene exons that function together during brain development. Exons are the parts of the genes that contain information coding for a protein. The group used an artificial intelligence algorithm graph clustering technique on gene expression data.

"This discovery was like finding a needle in a haystack, as there are thousands of variants in hundreds of genes thought to underlie autism, each of which is mutated in less than one percent of families with the disorder. We built a complex map, and then needed to develop a magnifier to zoom in," said Luo.

"The map and magnifier approach showcases a generalizable way of using multiple data modalities for subtyping autism and it holds the potential for many other genetically complex diseases to inform targeted clinical trials.”

Using the AI-enhanced tool, researchers also identified a strong association of parental dyslipidemia with autism spectrum disorder in their children. Moreover, the team saw altered blood lipid profiles in infants later diagnosed with autism spectrum disorder. These findings have pushed the research team to pursue further studies, including clinical trials that aim to promote early screening and early intervention of autism.

"Today, autism is diagnosed based only on symptoms, and the reality is when a physician identifies it, it's often when early and critical brain developmental windows have passed without appropriate intervention," said Luo. "This discovery could shift that paradigm."

The approach designed in this study is believed to be the first of its kind in precision medicine, and could pave the way for more personalized care and prevention for patients with autism.

"Previously, autism subtypes have been defined based on symptoms only -- autistic disorder, Asperger syndrome, etc. -- and they can be hard to differentiate as it really is a spectrum of symptoms," said Luo.

"The autism subtype characterized by abnormal levels identified in this study is the first multidimensional evidenced-based subtype that has distinct molecular features and an underlying cause."

Research teams have previously applied artificial intelligence and precision medicine techniques to autism research. A group at Princeton University recently used a deep learning algorithm to decode the functional impact of genetic mutations in people with autism.

The team was able to identify new mutations that could possibly contribute to the development of autism, which could help researchers understand how the condition differs from person to person.

“They say that when you meet one person with autism you have met one person with autism because no cases are alike. Genetically, it seems to be the same way,” said Chandra Theesfeld, PhD and lead author of the study.