TWIN-X

Multimodal Digital TWINs with Generative AI for eXplainable Precision Medicine

Description

The diagnosis and monitoring of patients with cancer and cardiovascular diseases, two of the leading causes of death in the European Union, depend on the integration of diverse health data. These include medical imaging, laboratory results, information on health risk factors, and clinical reports generated during patient consultations. Bringing together these diverse types of information and translating them into meaningful clinical insights is a complex and time-consuming process, often limiting the time physicians can spend in direct contact with their patients.

TWIN-X addresses this challenge by unlocking the potential of artificial intelligence. The project aims to develop a novel, trustworthy, and explainable AI framework that creates new opportunities for clinical decision support and medical research across Europe. Unlike conventional digital twin approaches, which typically merge all available patient data into a single mathematical representation, TWIN-X introduces a modular and hierarchical architecture that organises health information into clinically meaningful units. These modular expert AI models can be selectively activated and combined according to a specific clinical question.

By integrating multimodal health data into a comprehensive digital twin, the project will support clinicians in making more informed decisions for the diagnosis and treatment of cardiovascular and oncological diseases. Ultimately, this will contribute to strengthening Europe's digital health ecosystem and to the safe and effective integration of AI technologies into clinical practice.

Coordinator

PD Dr Keno Bressem

TUM Klinikum rechts der Isar

Email