Smart Mat for Respiratory Activity Detection: Study in a Clinical Setting Samuel Otis 1(B ) , Bessam Abdulrazak 2 , Soﬁa Ben Jebara 3 , Francois Tournoux 4 , and Neila Mezghani 1,5 1 Laboratoire de recherche en imagerie et en orthop´ edie, CRCHUM, Montreal, Canada samuel.otis.1@ens.etsmtl.ca 2 Department of Computer Science, Sherbrooke University, Sherbrooke, Canada Bessam.Abdulrazak@usherbrooke.ca 3 COSIM Laboratory, Carthage University, Higher School of Communications of Tunis, Ariana, Tunisia sofia.benjebara@supcom.tn 4 Department of Medicine, Centre Hospitalier de l’Universit´ e de Montr´ eal, Montreal, Canada francois.tournoux@umontreal.ca 5 Centre de recherche LICEF, Universit´ eT ´ ELUQ, Montreal, Canada neila.mezghani@teluq.ca Abstract. We discuss in this paper a study of a smart and unobtru- sive mattress in a clinical setting on a population with cardiorespiratory problems. Up to recently, the vast majority of studies with unobtrusive sensors are done with healthy populations. The unobtrusive monitoring of the Respiratory Rate (RR) is essential for proposing better diagnoses. Thus, new industrial and research activity on smart mattresses is tar- geting respiratory rate in an Internet-of-Things (IoT) context. In our work, we are interested in the performances of a microbend ﬁber optic sensor (FOS) mattress on 81 subjects admitted in the Cardiac Intensive Care Unit (CICU) by estimating the RR from their ballistocardiograms (BCG). Our study proposes a new RR estimator, based on harmonic plus noise models (HNM) and compares it with known estimators such as MODWT and CLIE. The goal is to examine, using a more represen- tative and bigger dataset, the performances of these methods and of the smart mattress in general. Results of applying these three estimators on the BCG show that MODWT is more accurate with an average mean absolute error (MAE) of 1.97 ± 2.12 BPM. However, the HNM estimator has space for improvements with estimation errors of 2.91 ± 4.07 BPM. The smart mattress works well within a standard RR range of 10–20 breaths-per-minute (BPM) but gets less accurate with a bigger range of estimation. These results highlight the need to test these sensors in much more realistic contexts. Keywords: Smart mattress · Ballistocardiogram · Respiratory rate This research was supported by the Canada Research Chair on Biomedical Data Mining (950-231214). c  The Author(s) 2019 J. Pag´an et al. (Eds.): ICOST 2019, LNCS 11862, pp. 61–72, 2019. https://doi.org/10.1007/978-3-030-32785-9_6