University Hospital Leipzig Releases Clinical Decision-Making System
University Hospital Leipzig relies on its own AI system "LAMPE" to support doctors in real-time diagnoses and therapies.
(Image: Shutterstock.com/ArtemisDiana)
With "LAMPE" – Leipzig Analysis and Reporting System for Patient Safety in Real-time – the University Hospital Leipzig (UKL) has integrated an AI-based Clinical Decision Support System (CDSS) into clinical practice. According to the UKL, it is the first fully self-developed system of its kind at a German university hospital that also meets the requirements of the European Medical Device Regulation (MDR).
In contrast to generative AI applications such as internal GPT systems, LAMPE is used directly in patient care. The system analyzes clinical data in real-time and supports doctors in diagnosis and therapy decisions with evidence-based recommendations for action, as stated in a press release from the UKL.
Videos by heise
A specific example of its use is the early detection of acute kidney failure in patients undergoing antibiotic therapy. Here, LAMPE continuously evaluates laboratory and treatment data and provides automated warnings to identify risks early on.
The system emerged from the multi-year AMPEL research project (Antibiotic Therapy and Patient Safety through Real-time Analysis of Laboratory Data) and was subsequently specifically developed into MDR-compliant software. For this purpose, the UKL established its own structures, including the Department of Medical AI and Translation (MedKIT) and a subsidiary for innovation transfer.
With LAMPE, the UKL pursues the goal of maintaining central digital tools under its own control and bringing them into care more quickly. At the same time, it is planned to make the underlying platform accessible to other clinics as an open-source solution in the future. An already established network with more than 20 locations is intended to promote professional exchange.
In the long term, the UKL sees systems like LAMPE as an important building block for more data-driven medicine, in which automated analyses relieve medical staff and create more time for direct patient care.
(mack)