Recurrence quantification analysis of photoplethysmography time series for assessing patients with peripheral artery disease

Abstract Peripheral artery disease (PAD) is the partial or complete occlusion of the major systemic arteries, and its diagnosis is commonly assessed using the ankle–brachial index (ABI). However, ABI is limited by arterial compressibility and is considered operator-dependent. Given the importance of early diagnosis for better outcomes and the complexity of PAD management, studying complementary tests and PAD’s mechanisms might help in its assessment. Photoplethysmography (PPG) is a low-cost, non-invasive technique that does not have arterial compressibility as a limitation and requires no advanced training to implement. The mean and standard deviation of PPG morphological parameters (pulse transit time, maximum systolic slope and pulse amplitude) have been studied before for PAD patients, but their non-linear dynamics have not been assessed. In this work, we performed recurrence quantification analysis (RQA) of PPG morphological parameters. A total of 40 legs from PAD patients with compressible ABI were studied for general description and to explore the relationship between PPG RQA indices and ABI. It was shown that RR interval dynamics differed from those of PPG-derived parameters despite their shared regulatory mechanisms. The central tendency and general variability of the PPG-derived parameters are related to changes in their dynamical states, which can increase or decrease their predictability, stationarity, or complexity. However, they do not clearly relate to PAD progression or PAD diagnosis. Further studies are needed, considering the results of this work and its limitations, to continue exploring the interesting PPG dynamics for a broader number of groups with different physiological conditions.