News
News from the BioCity Campus

When every hour counts: early detection of sepsis is possible with a small blood count – thanks to new AI-based methods

With new methods based on artificial intelligence (AI), the AMPEL project at Leipzig University Hospital is making significant progress in the area of ​​patient safety: now the small blood count is enough to detect patients with sepsis earlier than before. The well-known sepsis parameter procalcitonin can also be significantly exceeded.

The AMPEL - launched in 2018 as an analysis and reporting system to improve patient safety through real-time integration of laboratory findings (AMPEL) - has since developed further and is now considered a digital infrastructure that enables clinical AI applications in routine care.

Since it was launched over five years ago, the multi-award-winning project at the Leipzig University Hospital has supported nursing and medical staff in patient care by recognizing critical situations in real time and thus significantly increasing patient safety. Automated notifications improve the availability and weight of medical information. AMPEL uses simple calculations through to complex AI models and monitors all the necessary data live.

If sepsis, blood poisoning, is suspected, things can become critical and every hour counts. Survival of this often fatal disease depends largely on administering antibiotics as early as possible. The UKL's AMPEL project has now reached a milestone in early sepsis detection: Using new machine learning methods, the team was able to develop an AI model and scientifically confirm it at two further locations, in Germany and the USA.

“Our study on predicting sepsis using CBC was accepted by the world's leading laboratory medicine journal, Clinical Chemistry. “We show the potential of how AI methods and very few laboratory parameters that have already been collected can be used to implement constant screening for patients with the onset of sepsis in routine care,” explains Dr. Daniel Steinbach, doctor and research associate at the Institute for Laboratory Medicine, Clinical Chemistry and Molecular Diagnostics and at the UKL Data Integration Center and member of the AMPEL core team. “And the fact that our AI model significantly exceeds the prediction of the established marker procalcitonin (PCT) without additional costs should generate great interest.”

Laboratory values ​​of the small blood count are always available, but rarely used

The fact that the data from a small blood count could help to detect sepsis at an early stage initially came as a surprise, even to the interdisciplinary AMPEL team. “The initial results showed that the AI ​​models developed often use the data from the small blood count. These are laboratory values ​​that are always there but are rarely taken into account,” says Dr. Steinbach. According to the UKL expert, these laboratory findings hardly play a role in the clinical care of sepsis detection, even though they are determined in almost every laboratory test in every hospital. Specific laboratory parameters such as procalcitonin would be used.

Procalcitonin is a contradiction in terms, says Maria Schmidt, who is also part of the AMPEL core team: “Although it is used almost everywhere, studies regularly come to the conclusion that its predictive power is too low. What remains is a recommendation to guide the correct antibiotic therapy as the disease progresses.” In their sepsis study, they were now able to show that the informative value of procalcitonin can be significantly improved using machine learning methods and in combination with the small blood count, explains biometrician Schmidt.

The theory should now be followed by practical application: “In laboratory medicine, I don’t know of any AI model that has been tested as extensively and as extensively as the sepsis model we published. In the end, it remains just theory and only practical application shows whether and how great the support really is. But fortunately, that is exactly one of AMPEL’s core competencies,” says Dr. Steinbach.

Further development as an open source project

Martin Federbusch, a specialist in laboratory medicine, leads the AMPEL project and regrets that the UKL is still the only hospital in Germany with the kind of digital infrastructure that was able to be created by AMPEL. “So an important goal remains for us: transferring it to other locations,” emphasizes Federbusch. “Thanks to the broad support at our location, we are able to further develop the AI ​​platform AMPEL independently.”

Since this year, the AMPEL has been further developed as an open source project under the leadership of the Leipzig University Hospital by Prof. Toralf Kirsten and his department for Medical Data Science. The focus is on creating a non-profit AI infrastructure for healthcare that meets the highest standards of adaptability, interoperability and transparency.

Link to article in “Clinical Chemistry”: https://doi.org/10.1093/clinchem/hvae001

Press release from the “University Hospital Leipzig” from February 05.03.2024, XNUMX


previous news

Bone implants: Biomaterial helps with bone replacement

According to doctors at Hannover Medical School, a 3D-printed resorbable scaffold material from BellaSeno has successfully helped regrow a 14 cm long missing piece of bone in a large radius segment bone defect.

NextNews

MEDICAL FORGE supports startups through coworking laboratories

The MEDICAL FORGE Leipzig is a start-up program from biosaxony, the Saxon industry cluster in the fields of biotechnology, medical technology and healthcare. As a startup accelerator, they have been supporting aspiring new founders at the BIO CITY LEIPZIG technology center since April 2022, among other things by providing innovative manufacturing technologies such as HP's Jet-Fusion 3D printing.