A Study Unveils AI to Identify Autism

Due to the intricacy of the condition, diagnosing autism spectrum disorder (ASD) still presents a difficult issue that calls for highly qualified clinicians. Multifactorial neurodevelopmental disorder autism has a wide range of symptoms. The Centers for Disease Control and Prevention (CDC) estimate that one in 36 children in the United States has an autism diagnosis, yet there are no precise biochemical indicators for its detection. In a paper published in the journal Scientific Reports, Brazilian researchers suggest a quantitative diagnostic approach.

The study was based on the brain imaging results of 500 individuals, 242 of whom had an autism diagnosis. On the data, machine learning techniques were used.

“We began developing our methodology by collecting functional magnetic resonance imaging [fMRI] and electroencephalogram [EEG] data,” said Francisco Rodrigues, last author of the article. He is a professor at the University of São Paulo’s Institute of Mathematics and Computer Science (ICMC-USP) in São Carlos, Brazil.

“We compared maps of people with and without ASD and found that diagnosis was possible using this methodology,” Rodrigues said.

These maps were used by the researchers to train a machine-learning algorithm. The system was able to identify which brain abnormalities were connected to ASD with greater than 95% accuracy based on the taught instances.

The paper states that while much recent research suggests machine learning-based methods for diagnosing ASD, they only take into account a single statistical parameter and ignore brain network organization, which is the study’s key innovation. The connections between brain areas are displayed in brain maps or cortical networks. About 20 years ago, research on these networks first started, and it has provided a fresh perspective on neuroscience. “Just as a road with interruptions alters the traffic in a region, a brain with alterations leads to changes in behavior,” Rodrigues said.

The examination of fMRI data revealed alterations in specific brain areas connected to cognitive, affective, learning, and memory functions. When compared to controls, the cortical networks of those with ASD showed more segregation, less information dissemination, and less connectedness.

“Until a few years ago, little was known about the alterations that lead to the symptoms of ASD. Now, however, brain alterations in ASD patients are known to be associated with certain behaviors, although anatomical research shows that the alterations are hard to see, making diagnosis of mild ASD much harder. Our study is an important step in the development of novel methodologies that can help us obtain a deeper understanding of this neurodivergence,” Rodrigues said.

It will take years to put the methodology into practice; it is currently under development. Nevertheless, it will advance knowledge of cerebral distinctions and help doctors in the future, particularly in situations where a diagnosis is dubious.

Many different applications

According to Rodrigues, the study makes a negligible contribution to our knowledge of the connection between ASD and brain abnormalities. The implementation of this automatic diagnostic procedure requires a great deal of study. Apart from ASD, brain mapping can be helpful in the diagnosis of other diseases. Research from the past demonstrates that brain mapping can also be used to identify schizophrenia with high accuracy.

“We began developing novel methods to identify mental disorders a decade ago. We found that the diagnosis of schizophrenia can be much improved using brain networks and artificial intelligence. We also recently studied use of the methodology to investigate Alzheimer’s disease and found accurate automatic diagnosis to be possible,” Rodrigues said, referring to a study reported in 2022 in the Journal of Neural Engineering.

Small databases and the complexity of data collection are just two of the many difficulties that must be overcome, but as a general approach it can aid in the understanding of a number of ailments, and one of the group’s aims is to look into the connections between mental problems.

“How similar in terms of brain alterations are schizophrenia and Alzheimer’s? If we can find correlations between mental disorders, we may be able to develop novel medications and similar treatments for different conditions, or even adapt treatment for one condition to use in another. We’re a long way from this, but the path ahead is promising,” Rodrigues said.

The researchers anticipate that a deeper comprehension of how brain changes affect behavior would result in more effective public policy and humane and effective medical care. The interdisciplinary nature of the research being done on the topic makes the complexity of the topic clear. Members of the group came from facilities in Brazil, France, and Germany and comprised physicists, statisticians, doctors, and neuroscientists. They examined medical information gathered by neurologists, research on brain imaging conducted by neuroscientists, and algorithms created by physicists and statisticians, as well as other data.

The first author of the publication, Caroline Alves, conducted the investigation as part of her doctoral research. She has degrees in computer science, physical and biomolecular sciences, and physics.

Source Link