AI4TeenDep: Fighting Teenage Depression with AI

A growing global problem

Adolescent depression has become one of the most pressing mental health challenges worldwide. According to data from the World Health Organization (WHO), approximately 3.5% of young people between the ages of 15 and 19 suffer from depression, while in the 10-14 age group this figure reaches 1.4%. These statistics represent millions of young people facing significant difficulties during a crucial stage of their development.

Adolescence is a period of profound changes, both physical and emotional, and depression during this stage can have long-term consequences on personal, academic and social development. Early detection and appropriate treatment are critical to prevent future complications and improve the quality of life of these young people.

Our answer: AI4TeenDep

In Horus ML we have developed AI4TeenDep, an artificial intelligence based solution to identify patterns associated with depressive symptoms in adolescents. This project has been possible thanks to the valuable collaboration with IDavinci-Innovation and Cognitive Impulse Consulting.

Technology in the service of mental health

AI4TeenDep uses machine learning algorithms to process different types of data related to depression indicators:

1. Analysis of executive functions: We evaluate cognitive patterns related to planning, attention and working memory.
2. Voice and language processing: We analyze linguistic patterns, changes in voice tone and speech rhythm that may indicate depressive states.
3. Tailored interface: The application is designed with the specific needs of adolescents and mental health professionals in mind.

A project of international scope

Our tool is part of the international project #5G4PHealth, funded by CDTI through the CELTIC-NEXT EUREKA CLUSTER and coordinated by Artech International BVBA. This project seeks to develop and implement a 5G technology-driven platform that enables connected health services based on the P4 model (Predictive, Preventive, Personalized and Participatory) in multiple medical specialties.

AI4TeenDep functionalities

The application we have developed offers:

• Periodic non-invasive evaluations that can be completed by the adolescent in just a few minutes
• Monitoring of key indicators through natural interactions with the device
• Early warnings for professionals when worrying patterns are detected
• Trend visualization for easy identification of significant changes
• Personalized recommendations based on the user's individual profile

Our solution is designed to complement the professional assessment, providing additional information that can be valuable in the diagnostic and follow-up process.

The potential impact

AI4TeenDep aims to contribute significantly to the field of adolescent mental health:

• Facilitating early detection of risk indicators
• Providing objective data to support the clinical evaluation.
• Optimizing healthcare resources by helping to prioritize cases according to need
• Reducing barriers to help-seeking, especially relevant for adolescents

Looking to the future

This project represents for us an important step in the application of artificial intelligence to the field of mental health. We continue to work on the development and improvement of this tool, always with the ultimate goal of contributing positively to the health and well-being of adolescents at risk of depression.

We are deeply grateful to our collaborators for their invaluable contribution to this project and to the CDTI for their financial support through the CELTIC-NEXT EUREKA CLUSTER program. It is thanks to these synergies that we are able to advance in the development of technological solutions with real social impact.