SUMMARY
Researchers led by Drs. Maxime Cannesson and Eran Halperin at UCLA have developed a way to predict arterial blood pressure waveforms using EKG and SpO2 waveforms.
BACKGROUND
Monitoring and measuring vital physiological signals like arterial blood pressure (ABP) is critical for successful surgery. Ideally these types of signals should be measured continuously and non-invasively. However, many times these signals do not get monitored due to increased health risk introduced by obtaining the signal from a patient, additional time and resources required by health providers to set up monitoring, or prohibitive costs. The body comprises of many integrated systems, and many physiological signals can give insight or be used to predict other signals. The ability to impute or predict physiological waveforms from other waveforms allows healthcare providers to continuously monitor a patient in situations where the imputed waveform signal would otherwise be unavailable.
INNOVATION
Researchers from the Departments of Anesthesiology and Perioperative Medicine and Computer Science at UCLA have developed an algorithm that predicts or imputes physiological waveforms using other waveforms. As a proof of concept, their algorithm was able to predict ABP waveforms in real-time using EKG and SpO2 waveforms. This already provides benefits over how ABP is measured in the operating room: continuously using invasive techniques or non-invasively with measurements occurring intermittently. Along with the ability to monitor waveforms that wouldn’t be otherwise be there, imputed waveforms can also be used as input for other predictive algorithms, like assessing risk for developing complications or monitoring risk during a surgical procedure.
POTENTIAL APPLICATIONS
ADVANTAGES