Scientists Develop Emotion Detection AI to Detect Signs of Mental Illness in Time

According to reports, a new artificial intelligence has been designed to detect patterns associated with autism, schizophrenia and Alzheimer's disease by sifting through brain imaging data to discover people's mental health conditions, and the biggest advantage of this artificial intelligence is its ability to predict disease symptoms before they appear.

The AI model first uses brain imaging data from healthy adult test subjects and then shows them to patients with mental health problems, enabling the model to identify subtle changes that are imperceptible to the human eye.

The complex computer program was developed by a team of researchers at Georgia State University, and they suggest that one day it will be possible to detect signs of Alzheimer's disease in people as young as 40 years old, when the disease usually shows symptoms around age 65.

Based on functional magnetic resonance imaging (fMRI) technology, the AI matches brain imaging data sets from more than 10,000 people, which is known to measure brain activity by detecting changes in blood flow.

Once the AI system read the test subjects' underlying functional MRI data, the team fed the system a dataset of more than 12,000 people with autism, schizophrenia, and Alzheimer's disease, which served as a base database for the AI to easily identify patterns of signs of these three mental illnesses based on the test subjects' brain imaging data.

The team noted that using functional MRI to identify psychiatric disorders can be costly and staff must carefully comb through brain imaging data, but using an AI system substantially reduces cost and time.

Even if we know from other tests or family history that someone is at risk for a disease such as Alzheimer's, we still can't predict exactly when the disease will occur, according to study co-author Vince Calhoun, founding director of the TReNDS Center. Currently, we narrow the time window by capturing relevant disease sign patterns through brain imaging, which are present before the onset of the disease and are good disease predictors.