Signal Quality Assessment of Photoplethysmogram Signals Using Hybrid Rule- and Learning-Based Models

Photoplethysmography signals are crucial for a wide range of applications and, therefore, high-quality PPG signals are crucial to describe the cardiorespiratory status accurately. Motion artifacts can impair PPG-based applications, especially when these signals are recorded via wearable devices. Taking that in consideration, some researchers had proposed few methods for assessing the quality of these signals. Some rule- and learning-based approaches for PPG signal are available to determine the quality of the signal. In this paper, we propose a tradeoff between these two approaches by introducing a hybrid model that employs both learning and decision rules to determine the quality of the signal.