AI can diagnose mental health disease better than traditional tools

Credit: Unsplash+

A new study has found that an artificial intelligence (AI) assistant can talk with people about their mental health and offer more accurate assessments than the tools doctors usually use today.

This research brings hope for a better way to support mental health care using technology.

The study involved 303 participants, including people with confirmed mental health conditions like depression, anxiety, PTSD, autism, ADHD, and bipolar disorder. It also included a control group of people without any mental health diagnosis. Each person had an online conversation with an AI assistant named Alba.

Alba asked 15 to 20 open-ended questions about their mental health and then suggested possible diagnoses using a guide called DSM-5. DSM-5 is the standard manual that doctors use around the world to diagnose mental health conditions.

After the AI interview, the participants also completed standard rating scales for the nine most common psychiatric conditions. These rating scales are often used in clinics and hospitals, asking people to rate their symptoms using a fixed set of questions. The goal was to compare how well Alba did against these traditional tools.

The results were clear. Alba was more accurate than the rating scales for eight out of nine mental health conditions. Not only that, but Alba could better tell the difference between conditions that often look similar, like depression and anxiety.

The rating scales sometimes gave similar scores for both, making it harder to tell which condition a person might have. Alba, on the other hand, used the full interview to better understand the person’s experience and make a more specific suggestion.

Many participants said they had a positive experience with the AI assistant. They found Alba to be caring, relevant, and supportive. Some even said it felt easier to talk openly when answering Alba’s questions from the comfort of their home, without feeling judged.

Professor Sverker Sikström, a psychologist at Lund University and the leader of the study, said this kind of AI interview could be very helpful. He explained that having a safe and private conversation with an AI before seeing a doctor could help people feel more comfortable.

It could also save time for doctors and nurses by giving them a good first picture of the patient’s mental health. Importantly, this tool is not meant to replace psychologists or doctors. Instead, it could support their work by offering a first assessment that is both accurate and person-centered.

This research is different from earlier studies. Many previous digital tools could only assess one or two conditions, and they didn’t always follow proper medical guidelines. Alba, however, can consider all diagnoses in the DSM manual and explain why a certain diagnosis might fit. That’s a big step forward in how AI is used in mental health.

The study was published in the journal Scientific Reports and adds to the growing interest in using AI to improve mental health care. With continued research and development, tools like Alba may soon play a regular role in helping people get the right diagnosis and support faster and more accurately.

If you care about depression, please read studies about how dairy foods may influence depression risk, and B vitamins could help prevent depression and anxiety.

For more information about mental health, please see recent studies that ultra-processed foods may make you feel depressed, and extra-virgin olive oil could reduce depression symptoms.

The study is published in Scientific Reports.

Copyright © 2025 Knowridge Science Report. All rights reserved.