Respiratory Activity Classification Based
on Ballistocardiogram Analysis
Mohamed Chiheb Ben Nasr
1(B) , Sofia Ben Jebara
1 , Samuel Otis
2 ,
Bessam Abdulrazak
4 , and Neila Mezghani
2,3
1 Higher School of Communication of Tunis, Carthage University, Aryanah, Tunisia
{mohamedchihab.bennaser,sofia.benjebara}@supcom.tn
2 Laboratoire de recherche en imagerie et en orthop´ edie, CRCHUM,
Montreal, Canada
samuel.otis.1@ens.etsmtl.ca
3 LICEF Institute, TELUQ University, Montreal, Canada
neila.mezghani@teluq.ca
4 Department of Computer Science, Sherbrooke University, Sherbrooke, Canada
Bessam.Abdulrazak@usherbrooke.ca
Abstract. Ballistocardiogram signals describe the mechanical activity
of the heart. It can be measured by an intelligent mattress in a totally
unobtrusive way during periods of rest in bed or sitting on a chair. The
BCG signals are highly vulnerable to artefacts such as noise and movement making useful information like respiratory activities difficult to
extract. The purpose of this study is to investigate a classification method
to distinguish between seven types of respiratory activities such as normal breathing, cough and hold breath. We propose a feature selection
method based on a spectral analysis namely spectral flatness measure
(SFM) and spectral centroid (SC). The classification is carried out using
the nearest neighbor classifier. The proposed method is able to discriminate between the seven classes with the accuracy of 94% which shows
its usefulness in context of Telemedicine.
Keywords: Ballistocardiogram · Machine learning · Biomedical signal
processing · Spectral analysis
1 Introduction
The development of connected object for personalized services, especially for
monitoring purposes, have significantly increased worldwide over the last few
years [1]. More specifically those that deals with the monitoring of respiratory
and cardiac diseases. Indeed these diseases are among the leading cause of death
and disability in the world. One of these respiratory diseases is the Chronic
Obstructive Pulmonary Disease COPD [2] a progressive life threatening lung
disease. According to the World Health Organization [3], COPD affects more
than 250 million cases globally, a staggering 3.17 million deaths per year and
c
The Author(s) 2020
M. Jmaiel et al. (Eds.): ICOST 2020, LNCS 12157, pp. 79–88, 2020.
https://doi.org/10.1007/978-3-030-51517-1_7
on Ballistocardiogram Analysis
Mohamed Chiheb Ben Nasr
1(B) , Sofia Ben Jebara
1 , Samuel Otis
2 ,
Bessam Abdulrazak
4 , and Neila Mezghani
2,3
1 Higher School of Communication of Tunis, Carthage University, Aryanah, Tunisia
{mohamedchihab.bennaser,sofia.benjebara}@supcom.tn
2 Laboratoire de recherche en imagerie et en orthop´ edie, CRCHUM,
Montreal, Canada
samuel.otis.1@ens.etsmtl.ca
3 LICEF Institute, TELUQ University, Montreal, Canada
neila.mezghani@teluq.ca
4 Department of Computer Science, Sherbrooke University, Sherbrooke, Canada
Bessam.Abdulrazak@usherbrooke.ca
Abstract. Ballistocardiogram signals describe the mechanical activity
of the heart. It can be measured by an intelligent mattress in a totally
unobtrusive way during periods of rest in bed or sitting on a chair. The
BCG signals are highly vulnerable to artefacts such as noise and movement making useful information like respiratory activities difficult to
extract. The purpose of this study is to investigate a classification method
to distinguish between seven types of respiratory activities such as normal breathing, cough and hold breath. We propose a feature selection
method based on a spectral analysis namely spectral flatness measure
(SFM) and spectral centroid (SC). The classification is carried out using
the nearest neighbor classifier. The proposed method is able to discriminate between the seven classes with the accuracy of 94% which shows
its usefulness in context of Telemedicine.
Keywords: Ballistocardiogram · Machine learning · Biomedical signal
processing · Spectral analysis
1 Introduction
The development of connected object for personalized services, especially for
monitoring purposes, have significantly increased worldwide over the last few
years [1]. More specifically those that deals with the monitoring of respiratory
and cardiac diseases. Indeed these diseases are among the leading cause of death
and disability in the world. One of these respiratory diseases is the Chronic
Obstructive Pulmonary Disease COPD [2] a progressive life threatening lung
disease. According to the World Health Organization [3], COPD affects more
than 250 million cases globally, a staggering 3.17 million deaths per year and
c
The Author(s) 2020
M. Jmaiel et al. (Eds.): ICOST 2020, LNCS 12157, pp. 79–88, 2020.
https://doi.org/10.1007/978-3-030-51517-1_7