Can an artificial intelligence like ChatGPT detect early signs of depression?

Artificial intelligence algorithms like ChatGPT aren’t just impressive for what they’ve done with Bing or for being able to power virtual lawyers. Their technology is such that could generate a revolution in the humanized world of medicine and the newly popularized mental health. In fact, there are already studies that seek to establish whether this AI would be capable of detecting early signs of disorders such as depression.

Early detection of disorders such as depression is essential to ensure correct treatment and the well-being of the patient. However, it is often the least understood and difficult to reveal stage of the disease. This is why a group of researchers at the University of Victoria, Australia, they have set to work to integrate chatbots such as ChatGPT in the psychiatric process.

To do this, it has developed its own chatbot in the best ChatGPT style. On this occasion they have decided to call it DEPRA. With this tool, the researchers and authors of the preliminary study aim to establish whether an AI of this class could serve as support when detecting and helping patients in the earliest stages of the disease.

DEPRA is able to help participants examine their mental health status as a reference for medical professionals to help patients suffering from depression.

Study Early detection of depression using a conversational AI bot: A non-clinical trialpublished in Plos One

This ChatGPT-inspired AI could be the future for detecting depression and other disorders

gpt-4, ChatGPT

For the creation of the DEPRA chatbot, the researchers used Google’s Dialogflow platform. with its interface were able to build a kind of ChatGPT to diagnose depression in the selected sample population. In addition, the artificial intelligence was also able to filter the results based on their severity levels, based on the Hamilton Depression Scale.

See also  So you can put a custom background on the keyboard of the mobile device

DEPRA is integrated into Facebook Messenger, the platform used by the participants to connect with DEPRA. Subsequently, the chatbot proceeded to conduct a user survey, which took approximately 30 minutes to complete. In the study, the researchers note that interaction with artificial intelligence yielded “high levels of satisfaction” by the participants.

The chatbot’s artificial intelligence takes into account “mood, guilt, suicidality, insomnia, agitation or delay, anxiety, weight change, and bodily symptoms to diagnose levels of depression,” they describe. in the study. To put it to the test, a group of 50 Australian residents was recruited,

They indicated that the questions were easy to understand and answer, that it did not take as long as attending a face-to-face session with a psychiatrist, and that they preferred the option of texting through social media platforms compared to speaking with a psychiatrist. in a consultation session.

For researchers, the use of artificial intelligence to diagnose early depression has an added benefit. When conversing directly with a computer, patients may have fewer reservations when speaking, openly communicating their thoughts, something we’ve seen with ChatGPT as well. Likewise, the 24/7 availability of the tool also adds extra flexibility that does not exist in current methods with mental health professionals.

The promising results of DEPRA

Photo by Andrew Neel – Pexels

According to the results achieved by artificial intelligence, 30% of respondents did not show symptoms of depression. For this reason, the DEPRA chatbot decided not to recommend a session with a psychiatrist or other medical professional. However, 20% of the participants had very severe symptoms of early depression, while 22% showed moderate symptoms.

See also  WhatsApp optimizes the function of messages that can be seen only once

The research team plans to improve their DEPRA chatbot in the future. While artificial intelligence is currently based on a manual calculation based on the user’s response -to which it assigns a score-, a future version would take care of using machine learning algorithms and sentiment analysis to automatically calculate the level of depression, coming closer to technologies like ChatGPT. Thus, participants can have the results in a shorter time and receive the necessary help.

Yes indeed, Remember that this is a preliminary study.. There is still much more to study. Artificial intelligences like ChatGPT are relatively new, so our understanding of them is not that deep yet. However, they have already found ways to help us with important issues like health, and even predict breast cancer.

You may also like...