This research study's primary focus is to examine Neonatal Cardiopulmonary Resuscitation (NCPR) dynamics, particularly analyzing real-time Chest Compression (CC) dynamics. Specifically, the study evaluates the dynamics of CC depth, recoil, rhythm, force, bed height, hand position, and fatigue levels to improve the quality and delivery of CC.

In this investigation, various CC methods, including Two-Thumb (TT), Two-Finger (TF), and Cross Four-Finger (CFF), were employed to perform CC at three different bed heights: normal, extended, and short. The impact of these various methods and bed heights on fatigue levels and the quality of CC is thoroughly examined and analyzed.

Furthermore, the study introduces a real-time fatigue level predicting method, which has been developed to assess its influence on CC quality and its impact on medical practitioners. The predicting method was developed by evaluating three machine-learning algorithms: Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Artificial Neural Network (ANN). The results indicate that the Two-Thumb method at normal bed height is the most effective and efficient approach. Additionally, the real-time fatigue level predicting method demonstrates superior performance compared to human perception of fatigue.