Possible breakthrough in autism diagnosis. A multi-university research team co-led by Gustavo K. Rohde, professor of engineering at the University of Virginia, has Developed a system capable of identifying genetic markers of autism in brain images with 89-95% accuracy.
The finding raises the possibility that doctors could identify, classify, and treat autism and related neurological conditions without having to wait for behavioral signals. This could lead to more rapid treatment. “Autism is traditionally diagnosed based on behavior, but it has a strong genetic basis. A genetics-based approach could transform understanding and treatment,” the researchers wrote in a paper published in the journal Science Advances.
Rohde, a professor of biomedical, electrical, and computer engineering, collaborated with researchers at the University of California, San Francisco and the Johns Hopkins University School of Medicine, including Shinjini Kundu, a former graduate student of Rohde’s and the study’s first author.
While working in Rohde’s lab, Kundu, now a physician at Johns Hopkins Hospital, helped develop a generative computer modeling technique called transport-based morphometricsor TBM, which is at the core of the team’s approach.
Using a new mathematical brain modeling technique, The system reveals patterns of brain structure that predict variations in certain regions of the individual’s genetic codea phenomenon called “copy number variations,” in which segments of DNA are deleted or duplicated. These variations have been linked to autism.
TBM allows researchers to distinguish normal biological variations in brain structure from those associated with deletions or duplications. “Some variations are known to be associated with autism, but their connection to brain morphology, in other words, how different types of brain tissue such as gray or white matter are arranged in our brains, is not well understood,” Rohde said.Discovering how CNV relates to brain tissue morphology is an important first step to understand the biological basis of autism.”
How TBM cracks the code
Transport-based morphometry is different from other machine learning image analysis models because It is based on mass transport, which is the movement of molecules such as proteins, nutrients and gases into and out of cells and tissues..
“Most machine learning methods have little or no relationship to the biophysical processes that generate the data. Instead, they rely on pattern recognition to identify anomalies,” the researchers explained in their study. But Rohde’s approach uses mathematical equations to extract mass transport information from medical images, creating new images for visualization and further analysis.
Then, using a different set of mathematical methods, the system parses the information associated with autism-related CNV variations from other “normal” genetic variations that do not lead to neurological disease or disorder, what researchers call “confounding sources of variability.” Previously, these sources prevented researchers from understanding the “gene-brain-behavior” relationship, effectively limiting health professionals to behavior-based diagnoses and treatments.
If more appropriate mathematical models were used to extract this information, important discoveries could be made from this enormous amount of data. According to Forbes magazine, 90% of medical data is in the form of images, but there are no appropriate means to decipher them. Rohde believes that TBM is a master key. “From such enormous amounts of data, important discoveries could be possible if we used more appropriate mathematical models.”
#Autism #discovery #generic #code #deciphered #hope #early #diagnosis