This project deals with topics such as depression, anxiety, and ADHD. It also addresses the use of artificial intelligence in mental health diagnosis and treatment.
Ukraine: 7333 / Poland: 116 123 / Austria: 142 / Denmark: 70 201 201 Georgia: 770-408-0625 / Find help near you: findahelpline.com
By Sigrid Debois Reuss, Viktoriia Polshyna, Laura Wurm, Mehriban Karimova, Anastasiia Shpylka
A quiet crisis is unfolding in minds around the world.
Globally, there are 13 mental health workers per 100.000 people, according to WHO’s Mental Health Atlas. In 2019, one in eight people, around 970 million people globally, were living with a mental disorder, most commonly affective disorders like depression and anxiety. Since the COVID-19 pandemic, those numbers have surged by 26% and 28% respectively for anxiety and major depressive disorders in just one year, while access to proper care remains limited for many.
Now, the medical IT-field is racing to close the gap: with artificial intelligence. As AI begins to analyse voices, moods and even written thoughts, one question remains:
Do humans really trust AI with their innermost feelings?
Warsaw, 01.07.2025
of health care executives agree that newer technologies raise the credibility of health care providers.
It’s estimated that 1 in 3 women and 1 in 5 men will experience major depression in their lives.
Dr. hab. Mikołaj Pindelski, professor at SGH Warsaw School of Economics and an expert at the use of AI in corporate environments, is critical towards the idea of AI as a kind of “at-home therapist” critically – at least for now.
But would an AI-system really be able to detect sadness or diagnose a mental disorder just by listening to your voice or scanning your eye?
In South Korea, researchers recently trained an AI model to detect attention-deficit/hyperactivity disorder (ADHD) with 96 percent accuracy, simply by analysing the retina, the light-sensitive layer at the back of the eye. Elsewhere, AI is learning to interpret subtle shifts in voice tone, sleeping patterns or eye movement to flag early signs of depressive episodes.
Developed in Poland, MoodMon is an example of an AI-based tool designed to monitor and support one's mental health by detecting early signs of episodes within affective disorders, such as depression or bipolar disorder. It analyses data from users’ smartphones and wearables, including daily voice recordings, sleep patterns and activity levels, to identify emotional changes. The goal is to send timely alerts to patients, their therapists and caregivers to help prevent crises and support ongoing treatment.
Now, after eight years in development, the app is entering a new phase of clinical trials to further improve its accuracy and meet strict safety standards before being released for broader use. Crucially, MoodMon is not intended for casual or personal use, it can only be accessed through a psychiatrist’s prescription.
And MoodMon is just one example.
Across Europe and beyond, the medical IT field is evolving quickly, with artificial intelligence becoming an increasingly common tool in mental health care.
A 2025 study by Mental Health Europe explores how AI is already being used to monitor symptoms, support diagnosis and deliver personalized interventions for affective disorders such as depression and bipolar disorder. From chatbot-based therapy to tools that analyse sleep or voice data, the potential to close gaps is significant.
Only a human can understand another human.
Mariia Stetsiuk
The study by Mental Health Europe also raises critical ethical concerns. On a broader scale, researchers warn of systemic bias, loss of human-centred care and the danger that digital tools could shift attention away from necessary investments in public mental health infrastructure. The study further notes that existing frameworks like the European AI Pact are too general to fully address the specific risks AI poses in mental health care. It calls for targeted regulation that explicitly considers the emotional vulnerability of users and the long-term social impact of replacing human care with automated systems.