Algorithm-based arterial blood sampling recognition increasing safety in point-of-care diagnostics

Area of Science:

• Biomedical Engineering
• Clinical Monitoring
• Data Analytics

Background:

• Arterial blood pressure monitoring is standard in critical care.
• Accurate blood sample dating is crucial for patient management.
• Current methods for tracking blood withdrawal lack real-time precision.

Purpose of the Study:

• To develop and evaluate an algorithm for detecting blood withdrawal using real-time arterial blood pressure data.
• To enhance patient safety by minimizing sample mix-ups.
• To improve the accuracy of blood sample dating.

Main Methods:

• Real-time arterial blood pressure data from patient monitors were analyzed.
• An algorithm identified anomalies in systolic, diastolic, and mean pressures.
• Detection was based on pressure changes indicative of blood withdrawal manipulation.
• The algorithm was validated using data from animal studies in healthy pigs.

Main Results:

• The algorithm successfully detected 97% of blood withdrawals in retrospective analysis.
• It accurately identified 422 out of 434 blood withdrawal events.
• The system also detected 64 other sampling events for laboratory analyses.
• Achieved high performance metrics: sensitivity 0.97, precision 0.96, F1 score 0.97.

Conclusions:

• Arterial blood pressure monitoring data can reliably identify individual blood samplings.
• This method significantly enhances patient safety by preventing sample mix-ups.
• The real-time detection capability offers improved clinical workflow and data integrity.