Related Experiment Videos
Computerized ventilator data selection: artifact rejection and data reduction
W H Young1, R M Gardner, T D East
1Department of Medical Informatics, LDS Hospital, Salt Lake City, Utah, USA.
International Journal of Clinical Monitoring and Computing
|August 1, 1997
Summary
Automated data acquisition from ventilators is more efficient for settings but challenging for measured parameters. A three-minute median strategy is recommended for clinically acceptable artifact rejection and data collection.
Area of Science:
- Biomedical Engineering
- Respiratory Care Technology
- Medical Informatics
Background:
- Automated data acquisition from computerized ventilators is crucial for efficient patient monitoring.
- Manual charting of ventilator data is time-consuming and prone to errors.
- Establishing reliable automated strategies for data collection and artifact rejection is essential.
Purpose of the Study:
- To identify effective strategies for automated data acquisition from computerized ventilators.
- To determine acceptable methods for artifact rejection in ventilator data using the Medical Information Bus.
- To compare automated data collection with manual charting methods.
Main Methods:
- Surveyed medical practitioners to define clinically important ventilator events.
- Conducted a prospective study collecting data from 10 adult patients over 617.1 hours.
- Evaluated twelve different computerized data selection and artifact rejection algorithms.
Main Results:
- Automated collection of ventilator setting data (e.g., FIO2) was ~25% of manually charted data, showing increased efficiency.
- Measured parameters (e.g., tidal volume) exhibited high variability and numerous artifacts.
- Automated data capture generally increased data volume compared to manual charting (1.2x for 3-minute median).
Conclusions:
- Automated collection of ventilator setting data is straightforward and more efficient than manual methods.
- Automated selection and presentation of measured parameters pose significant challenges.
- Recommended strategy: record ventilator settings after 3 minutes and measured parameters using a 3-minute median to reject artifacts and ensure clinical acceptability.