Cardiovascular monitoring in sports via wearable PPG: a review on applications, accuracy challenges, and future opportunities

Purpose This paper aims to critically review and synthesize the current state of wearable photoplethysmography (PPG) technology for cardiovascular monitoring in sports. It also aims to provide a comprehensive analysis of its applications, the significant challenges to measurement accuracy and the emerging technological and analytical opportunities that are shaping its future. Design/methodology/approach A systematic literature review was conducted, drawing from over 100 peer-reviewed articles, conference proceedings and technical reports. The synthesis focuses on three core areas: the practical applications of PPG-derived metrics in athletic training and recovery; the multifaceted challenges to data fidelity, with a deep dive into motion artifacts and validation frameworks; and the evolution of signal processing and machine learning techniques for enhanced accuracy. Finally, the review explores future trends, including foundation models, multi-modal sensor fusion and ethical considerations. Findings Wearable PPG is widely applied for monitoring heart rate, heart rate variability and sleep, providing valuable insights for managing training load and recovery. However, its accuracy is severely hampered by motion artifacts (MAs), sensor location and subject-specific factors like skin tone and body mass index. While traditional signal processing methods offer partial solutions, deep learning (DL) models – particularly convolutional neural network, recurrent neural networks and generative architectures – demonstrate superior performance in MA mitigation. A new paradigm is emerging with foundation models trained on large-scale “in-the-wild” data, promising greater generalizability. Future progress hinges on multi-modal sensor fusion, advanced artificial intelligence and the establishment of robust validation and ethical frameworks. Originality/value This paper provides an exhaustive, up-to-date synthesis that connects the physiological applications of PPG in sports with the complex engineering and data science challenges of ensuring its accuracy. It offers a unique, critical perspective on the paradigm shift from traditional algorithms to advanced DL and foundation models, providing researchers and practitioners with a nuanced understanding of the field’s current state and future trajectory.