A Machine Learning-Based Pulse Detection Algorithm for Use During Cardiopulmonary Resuscitation

Abstract

Resuscitation guidelines mandate pausing chest compressions (CCs) during cardiopulmonary resuscitation (CPR) to check for the presence of pulse. However, interrupting CPR during a pulseless rhythm adversely affects survival. The aim of this study was to develop a pulse detection algorithm during CPR using the ECG and thoracic impedance (TI) signals. Data were collected from 116 out-of-hospital cardiac arrest (OHCA) patients during CCs and pulse/no-pulse annotations were carried out in artefact-free intervals by clinicians. CC artefacts were first removed from ECG and TI using recursive least-squares (RLS) filters. The impedance circulation component (ICC) was then derived from the filtered TI using a RLS-based adaptive scheme. The wavelet decomposition of the ECG and ICC was carried out to obtain the different subband components and the reconstruced ECG and ICC. A total of 124 discrimination features were extracted from those signals andfed into a random forest (RF) classifier that made the pulse/no-pulse decision. A repeated cross-validation procedure was used for feature selection, parameter tuning, and model assessment. Pulse/no-pulse diagnoses obtained through the RF were compared with the annotations to obtain the sensitivity (SE), specificity (SP) and balanced accuracy (BAC) of the method. The results obtained were: 76.2% (SE), 66.2% (SP) and 71.2% (BAC).