A Machine Learning-Based Blood Volume Classification Model for Cardiopulmonary Resuscitation Robot Feedback System

During Cardio Pulmonary Resuscitation (CPR), appropriate heart compression affects the quality of CPR, which is directly related to the patient’s life. Therefore, it is important to accurately judge the quality of CPR. Therefore, it is important to accurately judge the quality of CPR. Until now, there have been studies on bio signal-based CPR feedback systems such as \(\text {EtCO}_{2}\) (End tidal \(\text {CO}_{2}\) , \(\text {EtCO}_{2}\) ), Photoplethysmography (PPG). However, it is not possible to provide an accurate basis for improvement in compression. Therefore, in this study, a machine learning-based CBV (Carotid Blood Volume) classification model was developed for various bio-signal data. In the results, Sensitivity, Specificity, Precision, and Accuracy had values of 0.91, 0.97, 0.94, and 0.95, respectively, and showed high classification performance. Therefore, the CBV classification model presented in this study will be able to become a model based on a feedback system that can intuitively judge the quality of current CPR.