Machine learning based prediction of microbiological findings and infectious focus determination in adult intensive care patients

Status

Recruiting

Phases

Unknown

Study type

Observational

Source

DRKS

Registry ID

DRKS00039122

Enrollment

5000

Registered

2026-01-26

Start date

2026-04-15

Completion date

Unknown

Last updated

2026-08-03

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Acute infections of different localizations

Interventions

Group 1: In this retrospective, monocentric study, routine clinical data of adult patients treated between 2019 and 2024 in anesthesiology-led intensive care units at Heidelberg University Hospital ar

Eligibility

Sex/Gender

All

Age

18 Years to No maximum

Inclusion criteria

Inclusion criteria: Collection of at least one blood culture during the intensive care unit stay

Exclusion criteria

Exclusion criteria: Missing or incomplete data on blood culture results

Design outcomes

Primary

Measure	Time frame
As this is a retrospective, exploratory methodological study, there are no clinical endpoints in the classical sense. The primary endpoint of the study is the successful training, testing, and validation of a machine learning algorithm for the prediction of microbiological blood culture results. Model performance is assessed using the receiver operating characteristic (ROC) curve and the corresponding area under the curve (AUC) for the correct prediction of positive or negative blood culture results as well as the probable pathogen type.	—

Measure

Time frame

As this is a retrospective, exploratory methodological study, there are no clinical endpoints in the classical sense. The primary endpoint of the study is the successful training, testing, and validation of a machine learning algorithm for the prediction of microbiological blood culture results. Model performance is assessed using the receiver operating characteristic (ROC) curve and the corresponding area under the curve (AUC) for the correct prediction of positive or negative blood culture results as well as the probable pathogen type.

—

Secondary

Measure	Time frame
Secondary endpoints include an extended evaluation of model performance using additional metrics such as sensitivity, specificity, precision, recall, and F1 score. In addition, it is explored whether the inclusion of further microbiological findings (e.g., urine, tracheal secretion, stool, or wound cultures) can improve prediction accuracy regarding the infection focus and the presence of multidrug-resistant pathogens.	—

Measure

Time frame

Secondary endpoints include an extended evaluation of model performance using additional metrics such as sensitivity, specificity, precision, recall, and F1 score. In addition, it is explored whether the inclusion of further microbiological findings (e.g., urine, tracheal secretion, stool, or wound cultures) can improve prediction accuracy regarding the infection focus and the presence of multidrug-resistant pathogens.

—

Countries

Germany

Contacts

Public ContactMelanie Marhofer

Universitätsklinikum Heidelberg

melanie.marhofer@med.uni-heidelberg.de+4962215636254

Outcome results

None listed

Source: DRKS (via WHO ICTRP) · Data processed: Aug 10, 2026