Segmented-Beat Modulation Method-Based Procedure for Extraction of Electrocardiogram-Derived Respiration from Data Acquired by Wearable Sensors During High-Altitude Activity

Agnese Sbrollini1, Danilo Bondi2, Sofia Romagnoli1, Micaela Morettini1, Ilaria Marcantoni1, Tiziana Pietrangelo2, Vittore Verratti2, Laura Burattini1
1Università Politecnica delle Marche, 2Università degli Studi "G. d'Annunzio” di Chieti – Pescara


High-altitude sports are affected by hypoxic stress-related alterations, and, consequently, may trigger severe events such as sport-related sudden death. Lack of oxygen requires higher respiratory activity, reflecting into an increase of respiratory and cardiac rhythms; thus, into-the-field monitoring of respiration is essential. Athletes are used to use wearable sensors to monitor their activity, thus, these devices may be valid tool to support athletes' monitoring. The novel Segmented-Beat Modulation Method (SBMM)-based procedure was proposed to extract electrocardiogram-derived respiration (EDR) from electrocardiogram (ECG); thus, the aim of the study is to assess SBMM-based procedure for EDR extraction in data acquired by wearable sensors during high-altitude physical activities. Breathing rate series (BRS), respiration signal (RES) and ECG were recorded using BioHarness 3.0 by Zephyr from 3 expeditioners, while performing a trek up to 4,556m of altitude. EDR was extracted from ECG by SBMM-based procedure. BRS, RES, and EDR were segmented into 60-second windows and characterized in terms of breathing rate (BRBRS, BRRES, BREDR, respectively). BRBRS, BRRES, BREDR were compared by absolute difference, concordance correlation coefficient (CCC) and linear regression analysis. BRBRS values (Table) are lower than values of BRRES and BREDR (Table), which instead are similar. Difference between BREDR and BRRES (2[1;4]cpm) is lower than those computed between BRBRS and BRRES (8[3;14]cpm), and between BRBRS and BREDR (7[3;13]cpm). Moreover, a good agreement between BREDR and BRRES is confirmed by CCC (0.62, P<0.05) and regression line (BRRES=0.91∙BREDR+4.47cpm), differently from results obtained for BRBRS and BRRES comparison (CCC=0.27, P<0.05; regression line: BRRES=0.29∙BRBRS+29.15cpm) and for BRBRS and BREDR comparison (CCC=0.20, P<0.05; regression line: BREDR=0.18∙BRBRS+31.05cpm). In conclusion, SBMM-based procedure is a good method to extract EDR from data acquired by wearable sensors during high-altitude physical activities.