Comparison of recurrence plot features of swallowing and breath sounds

Nonlinear dynamics theory has been a tool of choice in the analysis of physiological signals and systems. This paper presents a novel approach in the analysis of tracheal swallowing and breath sounds based on nonlinear dynamics theory. As the tracheal sound signal is nonstationary, the signal was studied based on the recurrence quantification analysis (RQA) method, which is a useful technique in the analysis of nonstationary and noisy signals. Tracheal sound recordings of 15 healthy and 9 dysphagic subjects were studied. The multidimensional state space trajectory of each signal was reconstructed using Taken’s method of delays. The reconstructed trajectories were analyzed by the RQA technique. The preliminary results suggested that some recurrence parameters were appreciably different between swallowing and breath sounds. In order to confirm discriminating ability of the parameters, an automated method for extraction of swallowing sounds in the records of the tracheal swallowing and breath sounds was investigated. The swallowing sound extraction results were validated by manual inspection of the simultaneously recorded airflow signal and spectrogram of the sounds and also by auditory means. Experimental results proved that the investigated method more accurately detected the boundaries of swallowing sounds than methods proposed previously. Swallowing sound detection may be employed in a system for automated swallowing assessment and diagnosis of swallowing disorders (dysphagia) by acoustical means.