Continuous monitoring of cardiac output from TCG signals.

摘要

Continuous measurement of cardiac output (CO) is an important and difficult measure to obtain in an ambulatory environment. A novel ambulatory monitoring system (LifeShirt, VivoMetrics, Inc., Ventura, CA, USA) with three Inductive Plethysmographic (IP) sensors embedded in a garment, enables continuous monitoring of respiration from the ribcage and abdomen areas, and captures thoracocardiograph (TCG) signals from the thorax at the level of the left ventricle. This TCG signal provides a non-invasive measure of the volumetric contractions of the heart. The raw TCG signal must undergo extensive signal processing and digital filtering to extract a volume curve similar to the ventricular volume curve obtained through echocardiography. Typically the respiratory component has an amplitude of over twenty times that of the stroke volume curve. This investigation compares various signal processing algorithms such as spectral subtraction and adaptive filtering to separate these 2 components, which can occupy the same frequency band. These algorithms make use of the ribcage and abdominal signals to predict the respiratory component within the TCG signal. A dual axis accelerometer that measures posture and levels of activity aids filtering movement artifact. With the addition of a single lead ECG, ensemble averaging is used to smooth artifact in the signal, and CO may be obtained by including a heart rate measure. Additional measures can be derived including left ventricular systolic time intervals such as Pre-ejection period, Peak ejection rate and time to peak ejection rate. The results show that increases and decreases in SV and CO can be measured over time.