AI technology can use words to judge whether you have mental illness

Language is a fascinating communication tool that allows people to share ideas with each other. Often, if the clarity and accuracy of the language is used properly, the language will resonate with people. Language is also a tool for psychiatrists to assess a patient's specific psychiatric or mental disorders, including schizophrenia. However, these assessments often depend on the availability of trained professionals and the need for adequate facilities.

Join a team of IBM Research Computational Psychiatry and Neuroimaging Groups and University members around the world.

Together they developed an artificial intelligence (AI) that can predict the onset of a patient's mental illness with relative accuracy and overcome the above assessment obstacles. Their research on psychotic predictive AI has been published in the World Journal of Psychiatry.

Using the findings of the 2015 IBM study, the organization demonstrated the possibility of using AI to simulate differences in speech patterns in high-risk patients who later developed psychosis. Specifically, they use the AI ​​method called Natural Language Processing (NLP) to quantify the concepts of "poor speech" and "thinking flight" as syntactic complexity and semantic coherence.

Then their artificial intelligence evaluated the researchers' language mode of guiding them to talk about an hour.

“In our previous research, we were able to build a predictive model with a manual score of 80% accuracy, but the automation features reached 100%”, Principal Investigator and Computational Psychiatry and Neuroimaging Group at IBM Research Manager, telling futurism.

For their new study, the researchers evaluated a much larger group of patients engaged in different types of speech activities: talking about a story they just read. Cecchi said that by using their knowledge from the 2015 study to train their mental illness prediction AI, the team was able to build a retrospective model of the patient's speech model.

According to the study, this system predicts a final incidence of 83% for people with mental illness. If applied to patients from the first study, the AI ​​will predict the patient's final development as a mental glass AI contraction with 79% accuracy .

IBM researchers' mental illness prediction AI may eventually help mental health practitioners and patients. As Cecchi wrote in his 2017 IBM research report, the traditional method of assessing patients is very subjective. He and his team believe that the use of AI and machine learning as a so-called computational psychiatric tool can eliminate this subjectivity and increase the chances of accurate assessment.

This new study is only part of IBM Research's computational psychiatry research. As early as 2017, Cecchi's team and researchers at the University of Alberta conducted a study through the IBM Alberta Advanced Research Center. This special work combines neuroimaging techniques with artificial intelligence to predict schizophrenia by analyzing the patient's brain scans.

As for the new study, Cecchi believes that this may be an important step in providing neuropsychiatric assessment to the wider public, and improving the diagnosis of psychiatric morbidity may lead to improved treatment.

“This system can be used, for example, in clinics. Cecchi tells futurism that patients who are considered at risk can be classified quickly and reliably so that (always limited) resources can be used for those that are considered very People who may be mentally ill. People without a professional or clinic can send audio samples for remote assessment through Psychiatric Prediction AI.

Just as Cheki tells futurism, this approach is not limited to the spirit of glass. He said: "In other cases, similar methods can be used, such as depression. In fact, IBM researchers are already exploring the potential to calculate psychiatry. Helps diagnose and treat other diseases, including depression, Parkinson's disease and Alzheimer's disease, and even chronic pain.

Artificial intelligence is a true medical revolution. As these advanced systems reach the mainstream, we will enter a new era of medical care, hoping that this is an era in which anyone can get the best diagnosis and treatment options anywhere.