.png){kind=icon}
.png){kind=icon}
Return
Accuracy of Quantifying Hypotension During Surgery Using Physiological Sensor Data
.png){kind=icon}
OA
AI
DOI：10.1109/jtehm.2026.3698197.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Objective: During surgery it is common to measure the arterial blood pressure. One important reason is to monitor for hypotension, a too low blood pressure, which is known to be harmful. Since post-surgical complications correlate with hypotension, it is common to calculate the number of hypotensive events, their cumulative length, or their cumulative area under threshold. To make the data collection process manageable, parameter data is often stored only once every 15 seconds and then used to derive the amount of hypotension.Methods and procedures: In this study, we calculate the errors introduced in quantizing hypotension based on different amount of downsampling from 419 surgical patients at the open abdomen surgery unit at Karolinska University Hospital. Experiments with error mitigation are also tested. Results are compared with using the continuous blood pressure waveform data.Results: For a downsampling to once every 15 seconds, the relative bias was 27.0% for the number of hypotensive events, 30.0% for the cumulative hypotensive length, and 10.8% for the cumulative area under threshold. Error mitigation approaches often do not improve the errors.Conclusion: These results highlight the necessity of using high frequency data and that it is important to be aware of the errors when interpreting older studies involving quantification of hypotension and based on downsampled data.
Keywords:
Arterial blood pressure
downsampling
hypotension
mean arterial blood pressure
Journal
I
IF:
4.4
Papers:
574
Citations:
1.7K