Georgia: To prevent and treat debilitating diseases including: Alzheimer’s disease, crazyand autism, a new study at Georgia State UniversityTRENDS Center of can lead to early diagnosis.
The study was published in the Journal of Scientific Reports.
A team of seven Georgia State scientists developed a sophisticated computer program that was able to sift through large volumes of brain imaging data and find unexpected patterns related to mental health disorders. Functional magnetic resonance imaging (fMRI) scans used to generate brain imaging data assess dynamic brain activity by looking at minute variations in blood flow.
“We have built artificial intelligence models to analyze a significant amount of data from fMRI,” commented Sergey Plis, lead author of the study and associate professor of computer science and neuroscience at Georgia State University.
Rather than a snapshot like an X-ray or the more popular structural MRI, he likens this type of dynamic imaging to a movie, noting that “the data available is so much bigger, so much richer, than a blood test or a standard MRI.” But that’s the problem–making sense of that data. is difficult.
Additionally, fMRI is expensive and difficult to obtain in these specific situations. However, data mining can be performed on standard fMRIs using AI models. And there are plenty of them around.
According to Vince Calhoun, founding director of the TRENDS Center and one of the study’s authors, “huge datasets are accessible in individuals without a documented clinical problem.” By using these broad but disjoint public datasets, the performance of the model was enhanced on smaller, more focused datasets.
According to Calhoun, “new patterns have evolved that we can clearly link to each of the three brain diseases.”
A dataset with more than 10,000 individuals served as the initial training ground for the AI models as they were taught the basics of fMRI imaging and brain activity. Then, the researchers used a multi-site data set of more than 1,200 people, including people with Alzheimer’s disease, schizophrenia and autism spectrum disorders.
How does it work? It’s similar to how Facebook, YouTube, or Amazon start learning about you from your online behavior and start predicting your future behavior, likes, and dislikes. A computer program was able to determine the exact “moment” when the brain imaging data indicated an association with a relevant mental state.
For these findings to be clinically helpful they must be used before the disorder appears.
If we can identify risk factors for Alzheimer’s disease by age 40 and use markers to predict risk, we can take action, according to Calhoun.
As such, there may be ways to provide better or more efficient therapies if the risks of schizophrenia can be identified before actual changes in brain structure occur.
“We’re still unable to predict when it will develop, even if we know through previous tests or family history that someone is at risk for an Alzheimer’s-like disease,” Calhoun said. Brain imaging can shorten that window by detecting relevant patterns as they emerge before clinical disease becomes apparent.
According to Pliss, the idea is that after collecting a large imaging dataset, “our AI models will analyze it and tell us what they’ve discovered about specific diseases. We’re building systems to discover new knowledge that we’ve been unable to obtain. own.”
The study’s first author and a doctoral candidate in computer science at Georgia State University, Md. Mahfuzur Rahman said the aim of the study was to “connect big world and big datasets with small world and disease-specific datasets and move towards relevant markers. Clinical decisions.” (ANI)
