Using empirical mode decomposition to increase the sensitivity of diode laser-based detection of ¹³CO₂ in human exhaled breath

Abstract

This paper reports on using the empirical mode decomposition of an experimental signal to improve the accuracy of 13CO2 detection in expired air through an absorption spectroscopy method that employs a frequency-tunable diode laser operating near a wavelength of 2 µm. The conducted research shows that using this method can increase the sensitivity of 13СO2 detection in human exhaled breath by two orders of magnitude.

Implementation of the adaptive Wiener filtering algorithm in the problem of measuring the 13CO2 content in expiratory air using the tunable diode laser absorption spectroscopy technique

“…It should be noted that a similar situation has been observed in [3] when using the Kalman filter, and the level of residual noise was about 5 times higher when using the Wiener filter compared to the results obtained using the Kalman filter.…”

“…One of the effective approaches for solving the problem of the highly accurate on-line measurement of 13 CO 2 content, in a mixture with 12 CO 2 in a human expiration for a non-invasive diagnosis of gastrointestinal cancer, is implementing the tunable diode laser absorption spectroscopy (TDLAS) technique using a tunable diode laser (TDL) as a source. This emits nearly a wavelength of 2 µm, where 13 CO 2 and 12 CO 2 have a number of weakly overlapping absorption lines, as noted earlier in a number of publications (see [1][2][3][4][5][6][7][8][9][10], for example).…”

“…In this paper, as before in [3,4], the Wiener filter has been used for processing an experimental signal obtained in [5]. It investigated the spectral range 4876.5-4879 cm −1 , in which the 13 CO 2 molecule had an absorption line P(12) and the 12 CO 2 molecule-in the absorption lines R (32) and R (34).…”

“…Other filters have been applied to the result obtained using the Wiener filter-Savitsky-Golay (SG), FFT, EMD, and EMD followed by applying the SG filter to the result, as in [3]. The results are presented in figure 2.…”

“…We have previously reported on the application of the EMD algorithm [3] and the Kalman filter [4] for filtering an experimentally recorded signal to improve the 13 CO 2 content measuring accuracy in human expiration, using a TDLAS technique based on the TDL at a wavelength of approximately 2 µm. This has established in these works that the application of these adaptive filtering techniques to the original signal allows one to obtain the measuring accuracy of 0.4% for 13 CO 2 content in a mixture with 12 CO 2 in expiratory air [3] and 0.15% [4], that is two or more than five times better, respectively; compared with the results obtained while processing the original signal using a non-adaptive fast fourier transform filter (FFT) [5].…”