Not long ago, AI in EU-funded health projects was mostly treated as a tool: an algorithm added to an app or a model trained to spot patterns in data and therefore improve diagnosis. Now, AI can also help people understand their health and help them stay well after cancer treatment ends.

Global Health Connector is a partner in two EU-funded projects that use AI in different ways.

CURTAIN: using trustworthy AI to tackle cancer inequalities

The CURTAIN Project is a new three-year initiative co-funded by the EU4Health Programme and coordinated by the Center for Innovation in Medicine – INOMED.

By adopting the acronym CURTAIN (Cancer ineqUalities thRough liTerAcy IN Europe), the project highlights the persistent cancer inequalities between Eastern and Western Europe, which are visible today through the lens of the former Iron Curtain.

To close that gap, CURTAIN focuses on cancer literacy among citizens, health professionals, policymakers, researchers and businesses. By bringing these groups together, the project aims to improve cancer prevention, diagnosis, care and survivorship across Europe. Partners will run training programmes, Living Labs, workshops and awareness campaigns, and will develop new digital health tools. One of these is a systematic, evidence-based framework for the ethical, safe and effective use of Large Language Models (LLMs) in cancer literacy. As more patients turn to AI chatbots for health information, this kind of guidance is becoming essential.

MAYA: an AI-powered health hub to improve the quality of life of cancer survivors

The MAYA Project is a four-year initiative funded by Horizon Europe and coordinated by the University of Ioannina.

For Adolescent and Young Adult (AYA) survivors, the late effects of treatment can raise the risk of cardiovascular disease years later. Care often falls through the cracks just as patients move out of active treatment.

To fill this gap, MAYA partners are building the iCARE health hub. It is a smart mirror connected to wearables (a smartwatch, blood pressure monitor, ECG and smart scale) that helps survivors monitor and manage their heart health at home. It has three AI components:

1. A conversational health assistant. Users can talk or type to it. A “supervisor agent” works out what the user needs and routes the question to a specialist: a HealthRAG agent for health literacy and lifestyle questions, or a physio-exercise agent. It runs on Azure OpenAI GPT-4o-mini, with Azure speech-to-text and text-to-speech.
2. A virtual exercise coach. Computer vision tracks the user’s posture through skeletal keypoints, counts repetitions and gives real-time feedback on form. Programmes start at group level and become more personalised over time.
3. Smart ECG analysis. Signal processing and a supervised machine-learning model tell normal heartbeats apart from abnormal ones, flagging conditions such as arrhythmias or left ventricular hypertrophy.

The iCARE hub is now being tested in a controlled trial with 300 participants across five pilot sites in Germany, Italy, Poland, Portugal and Spain. The primary outcome is quality of life, which is a reminder that the goal of this AI is to help survivors live better, not only to collect data.

CURTAIN and MAYA focus on different stages of the cancer journey, but they point to a similar challenge: AI in healthcare does not work in isolation. Its impact depends on how well it is integrated into care, how people are supported in using it, and whether it is developed with ethical considerations and patients’ needs in mind.

This is where collaboration across research, industry, healthcare and patient communities becomes important. At Global Health Connector, we work to bring these different perspectives together so that new technologies are not only developed, but also understood, trusted and used in practice.

Follow the CURTAIN and MAYA projects to discover how this work is progressing.