Artificial Intelligence can detect autism much earlier than when the child initiates the interaction
2024-09-04 09:14:03
TiranaPost
The genetic code of autism has been deciphered, thanks to AI. Detection is of high importance as it can speed up the time to diagnosis.
Early diagnosis would therefore lead to earlier treatment of autism.
The discovery was made by a team of researchers that includes several universities and who have realized this using Artificial Intelligence. The entire research is described in detail on the pages of the specialized journal "Science Advances".
Researchers have developed an AI-based system capable of identifying genetic markers of autism through biological activity in the brain with 89-95% accuracy. According to the authors, this approach could soon revolutionize the diagnostic process, significantly reducing the waiting time for families.
The new method developed by the researchers, co-led by Gustavo K. Rohde, professor of Engineering at the University of Virginia, is based on a combination of brain mapping and artificial intelligence. First, brain images are taken by MRI. These images are then analyzed by AI, which can detect the movements of specific proteins, nutrients and other biological processes in the brain, signals that may indicate the presence of autism. Currently, the diagnosis of autism is usually made only after signs of unusual behavior are clearly manifested.
"A genetics-based approach could transform the understanding and treatment of autism," the researchers underlined.
