Umeå University joins EU project to improve doctor–patient interaction

ASCLEPIUS has received €20 million in funding through the Horizon Europe call GenAI4EU. The project brings together researchers, clinicians, patient representatives, ethicists and social scientists from 17 partner organisations in seven countries. At Umeå University, the work is led by Professor Frank Dignum, Director of TAIGA – Centre for Transdisciplinary AI for the Good of All.

AI support for better patient-clinician communication

Healthcare systems across Europe face growing challenges. As demand for care increases, doctors and other healthcare professionals spend a substantial amount of their time on documentation, information management and administrative tasks.

ASCLEPIUS will develop virtual AI assistants designed to support both patients and healthcare professionals throughout different stages of the care journey. The assistants are meant to strengthen clinical judgement, not replace it, so that people can make better-informed decisions.

Umeå University's role in the project is to contribute theory and tools for building interaction support that fits naturally into the conversation between doctor and patient, improving both quality and efficiency.

From 2027 to 2030, researchers will explore how AI can be applied in areas including rehabilitation, intensive care, neurology and internal medicine.

Frank Dignum, Professor and Director of TAIGA

Image Mattias Pettersson

"ASCLEPIUS is exactly the kind of transdisciplinary AI research that TAIGA promotes and that will actually have a positive impact on how medicine in hospitals can be practiced. It will benefit both doctors, patients and other health care workers involved in patient care", says Frank Dignum.

Keeping people at the centre

A key principle of the project is that AI systems must be understandable, transparent and trustworthy. They should be able to explain their recommendations, communicate uncertainty and adapt to different cultures and healthcare environments. The tools will be available in the languages of the countries where they are used: Dutch, Portuguese, French and English.

Imagine a medical consultation where, after the appointment, the clinician receives help summarising the key points and documenting them in the patient's record. At the same time, the patient receives a clear summary of what was discussed and what happens next.

Parts of this are already possible with today's generative AI. What is still missing is a way to ask follow-up questions once the summary exists, and a way for the clinician's corrections to reach everyone with access to the record, including the patient. By taking over parts of the information management, AI could free up more time for the human side of healthcare. Exactly what these solutions will look like is something the researchers will develop and evaluate during the project.

Building on the TAIGA approach

For TAIGA, the project also highlights the value of bringing together researchers from different disciplines to address shared AI-related challenges. ASCLEPIUS builds on the same idea: that the best technology is developed together with the people who will actually use it. According to Frank Dignum, TAIGA has played an important role in Umeå University's participation in the international consortium.