Development of a machine learning model for the prediction of patient-specific piperacillin serum concentration range

Status

Active, not recruiting

Phases

Unknown

Study type

Observational

Source

DRKS

Registry ID

DRKS00033743

Enrollment

1000

Registered

2025-04-28

Start date

2025-10-24

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

Adherence to the desired serum concentration range during antibiotic therapy with Piperacillin/Tazobactam

Interventions

Group 1: Patients who have received an antibiotic therapy with Piperacillin/Tazobactam as part of an intensive care treatment. The primary aim of the study is to develop a machine learning model that

Eligibility

Sex/Gender

All

Age

18 Years to No maximum

Inclusion criteria

Inclusion criteria: Antibiotic therapy with Piperacillin/Tazobactam, Piperacillin/Tazobactam blood level measurement performed

Exclusion criteria

Exclusion criteria: If an individual patient dataset proves to be insufficient for further analysis and model training after initial analysis of the entire dataset, it will be removed

Design outcomes

Primary

Measure	Time frame
In this study, there is no primary endpoint in the conventional sense. The primary goal of the study is the completed development of a machine learning model that has been successfully tested and validated within the scope of our dataset. After selecting a suitable model architecture, the model will be trained using the prepared patient data. Approximately 70% of the available data will be used to build the model. Another 15% or so will be used to test the newly developed system. Finally, the model will be tested on a separate validation dataset comprising the remaining 15% of the data to ensure that it generalizes well to unseen data. Based on the validation results, the model will be further optimized and refined. All common state-of-the-art machine learning algorithms (e.g., Random Forest) will be used, and their results will be compared using appropriate statistical metrics (e.g., area under the receiver operating characteristic curve (AUC), receiver operating characteristic curve (ROC)). All analyses and model training will be carried out using Python software version >3.8.0.	—

Measure

Time frame

In this study, there is no primary endpoint in the conventional sense. The primary goal of the study is the completed development of a machine learning model that has been successfully tested and validated within the scope of our dataset. After selecting a suitable model architecture, the model will be trained using the prepared patient data. Approximately 70% of the available data will be used to build the model. Another 15% or so will be used to test the newly developed system. Finally, the model will be tested on a separate validation dataset comprising the remaining 15% of the data to ensure that it generalizes well to unseen data. Based on the validation results, the model will be further optimized and refined. All common state-of-the-art machine learning algorithms (e.g., Random Forest) will be used, and their results will be compared using appropriate statistical metrics (e.g., area under the receiver operating characteristic curve (AUC), receiver operating characteristic curve (ROC)). All analyses and model training will be carried out using Python software version >3.8.0.

—

Countries

Germany

Contacts

Public ContactJan Larmann

Universitätsklinikum Aachen Klinik für Anästhesiologie

jan.larmann@med.uni-heidelberg.de+492418088179

Outcome results

None listed

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