The AI application can be operated in the so-called Cloud Edge continuum directly in the hospital on its own computers without a connection to the internet or via the cloud. The data can thus be strictly protected and only stored in Europe in accordance with European data security standards. The one-year project to develop a prototype based on shock room simulations started in September. The basis is a modular software kit for the AI solutions.
How AI should support the shock room and facilitate documentation
The medical team in the shock room is under immense pressure: all medically relevant information is communicated verbally during the handover of patients from the ambulance service to the shock room team, during the treatment process in the shock room and when patients are handed over to the intensive care unit or the operating theatre. Information has to be recorded, exchanged and processed at lightning speed, while diagnostics and treatment take place in parallel. The solution to be developed uses artificial intelligence to automatically record the conversations, evaluate them in real time, enrich them and prepare them in a structured graphic format. The AI agent currently under development is designed to help with this.
Structure for emergencies: the ABCDE scheme saves lives
Treatment follows the so-called ABCDE scheme, with life-threatening conditions being treated first: Airways (Airways), Ventilation (Breathing), Circulation (Circulation), Neurological Deficit (Disability), Advanced Information (Exposure). Are the airways clear, is there severe internal or external bleeding, is the metabolism derailed? But also: Is the injured person taking anticoagulants and therefore likely to bleed to death more quickly? The aim of AI is as follows: The AI recognizes which findings, measures and decisions are mentioned. If the doctor mentions a “coarse rattling sound” during breathing, the AI agent determines the category and creates a live image in traffic light logic according to the ABCDE scheme. In addition, the AI agent automatically transfers the data to the forms for documentation and quality assurance.
Model case for secure and flexible cloud edge infrastructures
The solution is currently being developed as part of the European funding program IPCEI-CIS. The aim is to create a standardized, transferable infrastructure in the cloud edge continuum. The system has considerable resilience to infrastructure failures because it can be operated both locally and offline via the DGX Spark super-minicomputer from NVIDIA as well as via the AI Foundation Services in the Open Telekom Cloud (OTC). The OTC is a key component of Deutsche Telekom’s T Cloud offering and complies with all European data protection regulations. Key components of the system such as the modular software construction kit for AI solutions, the edge agent framework, the lean AI models and automated workflows for data processing and training can be “reused” in the development of other AI agent-based cloud edge infrastructures.
Progress through a strong partnership
“By developing a versatile multi-agent framework and adapting it to the requirements of emergency medicine, we are creating the basis for relieving the burden on staff in the care of seriously injured patients. Important building blocks for this are a well thought-out system architecture, the integration of trustworthy components for voice processing and data management as well as edge capability for operation directly on site,” says Stefan RĂĽping, Head of Department at Fraunhofer IAIS.
“By contributing medical expertise and realistic shock room simulations, we are interlinking research and practice and helping to ensure that AI agents in the shock room could bring tangible progress for emergency care in the future,” adds Jerome Defosse, Senior Consultant at the Clinic for Anesthesiology and Surgical Intensive Care Medicine at the Cologne City Hospitals. “AI is already helping to save lives today. Our AI agent for emergency medicine is also a role model for other sectors. With this practical solution, we are demonstrating the benefits of sovereign digital infrastructure for the economy and the common good,” says Ferri Abolhassan, CEO of T-Systems and member of the Board of Management of Deutsche Telekom AG.
The Bundeswehr Hospital in Berlin and the Florence Nightingale Hospital in DĂĽsseldorf are supporting the project in an advisory role. A prototype running on site in the hospital, which is also fully functional offline, is expected to be ready for use in summer 2026. Thorsten Tjardes, Director of the Clinic for Trauma Surgery and Orthopaedic Septic and Reconstructive Surgery at the Bundeswehr Hospital Berlin: “For us, the resilience of the AI assistance has top priority – it must also function absolutely reliably and safely in the cloud and locally, i.e. offline, in order to meet the high requirements of emergency medicine.” Martin Pin, Head of the Clinic for Emergency and Acute Medicine at Florence Nightingale Hospital in DĂĽsseldorf, adds: “As a consulting partner, we are contributing our expertise in order to design innovative AI technology in a practical way and ensure an optimal balance between technical excellence and usability in clinical operations.”
The current project builds on the TraumAInterfaces project funded by the German Federal Ministry of Health. In order to further develop the knowledge gained there and make it available to the healthcare sector, follow-up funding was provided by the European funding program IPCEI-CIS.
The AI Foundation Services are a central platform for secure AI applications. Companies can find a large selection of open source and commercial AI models there. The open source models are operated by T-Systems according to the highest security standards in its own data centers.
The EU’s IPCEI-CIS program aims to create a “multi-provider cloud edge continuum”: A networked, sovereign digital infrastructure for Europe by Europe. Edge cloud computing places powerful computing capacities at the edge of the network with minimal latency. Twelve EU member states and around 150 partners, including industry giants such as SAP, Siemens, Bosch, TelefĂłnica, Orange and Airbus, are already working on this open, European operating system.
About T-Systems
With locations in 26 countries, more than 26,000 employees and annual revenues of over 4 billion euros, T-Systems is one of the leading service providers for information technology and digitalization solutions in Europe.
About Fraunhofer IAIS
The Fraunhofer Institute for Intelligent Analysis and Information Systems IAIS, based in Sankt Augustin near Bonn with locations in Dresden and Heilbronn, is one of the leading institutions for applied research in the fields of artificial intelligence (AI), machine learning and generative AI in Germany and Europe. The approximately 350 employees develop strategies, technologies and solutions for companies, authorities and organizations along the entire value chain.
About Kliniken der Stadt Köln gGmbH
Kliniken der Stadt Köln gGmbH is one of the largest municipal healthcare providers and is a university hospital of the private University of Witten / Herdecke. Around 159,500 outpatients and 48,500 inpatients are treated each year. The hospital has 1,400 beds and employs around 5,000 staff at its three sites in Cologne-Merheim, Cologne-Holweide and Amsterdamer StraĂźe Children’s Hospital. In future, the three sites will be merged to form a health campus at the Merheim site. Klinikum Köln Merheim is certified as a supra-regional trauma center by the German Society for Trauma Surgery. Every year, 600 patients are treated in the hospital’s trauma rooms. As part of the project, the clinics of the city of Cologne are subcontractors of Fraunhofer IAIS.
– – – – – –
Further links
👉 www.iais.fraunhofer.de
Image: Image created with GPT-5
