Pediatric Pulmonology 46:523–529 (2011)

State of the Art

Respiration Rate Monitoring Methods: A Review

F.Q. AL-Khalidi,1 R. Saatchi,1 D. Burke,2 H. Elphick,2* and S. Tan1
Summary. Respiration rate is an important indicator of a person’s health, and thus it is monitored
when performing clinical evaluations. There are different approaches for respiration monitoring,
but generally they can be classed as contact or noncontact. For contact methods, the sensing
device (or part of the instrument containing it) is attached to the subject’s body. For noncontact
approaches the monitoring is performed by an instrument that does not make any contact with the
subject. In this article a review of respiration monitoring approaches (both contact and noncontact)
is provided. Concerns related to the patient’s recording comfort, recording hygiene, and
the accuracy of respiration rate monitoring have resulted in the development of a number of
noncontact respiration monitoring approaches. A description of thermal imaging based and vision
based noncontact respiration monitoring approaches we are currently developing is provided.
Pediatr Pulmonol. 2011; 46:523–529. ß 2011 Wiley-Liss, Inc.

Key words: respiratory; thermal imaging; noncontact.

Funding source: The Children’s Hospital Charity

INTRODUCTION the acute medical admissions unit, respiratory rate was

only measured in around two thirds of children on arrival.
Breathing is an important physiological task in living
Reasons for omission were given as ‘‘crying’’ or ‘‘unset-
organisms. For humans, this process results in air contain-
tled’’ and in many of these children experienced nursing
ing oxygen being inhaled into the lungs, where gas
staff were unable to obtain an accurate reading. However,
exchange occurs across the alveolar-capillary membrane.1
it is in these children that this measurement would poten-
Carbon dioxide is excreted as part of the process, in the air
tially be most useful and important.
released through the nose or mouth. The entire process
According to Murthy et al.6 existing devices for
from the inhalation to exhalation is known as a breathing
monitoring respiratory rate only estimate the actual
(or respiration) cycle.
breathing rate due to their limitations. These devices
Respiratory rate is a vital sign used to monitor the
can be classified in different ways, depending on the
progression of illness and an abnormal respiratory rate
manner in which they operate and how they are used.
is an important marker of serious illness. There is sub-
In this article, respiration rate monitoring devices are
stantial evidence that alterations in respiratory rate can be
used to predict potentially serious clinical events such as
cardiac arrest or admission to the intensive care unit.2–5
These studies have shown respiratory rate to be better than 1
Faculty of ACES, Sheffield Hallam University, Sheffield, UK.
other vital measurements such as pulse and blood pressure
2
in discriminating between stable patients and patients at Sheffield Children’s NHS Foundation Trust, Sheffield, UK.
risk.3 Using changes in respiratory rate measurements
*Correspondence to: Dr. H. Elphick, Floor E, Stephenson Unit, Sheffield
patients could have been identified as high risk up to Children’s Hospital, Western Bank, Sheffield S10 2TH, UK.
24 hr before the event with a specificity of 95%.5 Hospital E-mail: [email protected]
systems such as the Pediatric Early Warning System
(PEWS) have been developed to encourage appropriate Received 11 July 2010; Revised 19 November 2010; Accepted 19 November
responses to abnormal vital signs including an elevated 2010.
respiratory rate. DOI 10.1002/ppul.21416
Evidence from a recent audit of febrile children in our Published online 31 January 2011 in Wiley Online Library
unit suggests that in both the emergency department and (wileyonlinelibrary.com).

ß 2011 Wiley-Liss, Inc.

524 AL-Khalidi et al.

grouped as either contact or noncontact. In contact respir- airflow, but the method is limited due to a high incidence of
ation rate monitoring, the instrument makes direct contact thermistor displacement.11
with the subject’s body. However, in noncontact monitor- The nasal pressure transducer is another sensor used to
ing, the respiration rate is measured without the instrument measure respiration rate. Nasal pressure is a more accurate
making contact with the subject’s body. There are clear measure of airflow than others as it based on the actual
advantages to noncontact respiration monitoring methods. volume of the air exhaled.8,12 It can be measured via nasal
These include improved patient comfort (especially for cannulae, mouthpiece, or facemask. A problem with air-
long term monitoring) as the subject is not tied to an flow measurement is that some patients may not feel
instrument and improved accuracy as distress caused by comfortable with the sensor.13 The collector can also
a contact device may alter the respiration rate. affect respiratory activity by increasing deadspace.
There have been several studies reporting developments Folke et al.14 have reported a CO2 sensor to measure
in both contact and noncontact respiration monitoring. A respiration rate. Their study also indicated that subtle
review of non-invasive respiratory monitoring in medical design changes in the collecting device could introduce
care was provided by Folke et al.7 This article aims to large differences in sensor performance.
review the literature on contact and noncontact methods of
respiratory rate monitoring, including two methods cur- Chest and Abdominal Movement Detection
rently being developed by our unit.
Chest and abdominal wall movements can best be
measured by either mercury strain gauges or impedance
CONTACT BASED RESPIRATION MONITORING methods. Respiratory inductance plethysmography is a
non-invasive technique whereby two bands measure the
Contact respiration rate monitoring instruments are
respiration rate, the thoracic band which is placed around
usually based on measuring one of the following para-
the rib cage and the abdominal band which is placed over
meters: respiratory sounds, respiratory airflow, respiratory
the abdomen at the level of the umbilicus. The bands are
related chest or abdominal movements, respiratory CO2
made from an extendible/deformable conducting material,
emission and oximetry probe SpO2. Respiration rate can
either a very fine wire or thin foil such that the conductivity
also be derived from the electrocardiogram (ECG).
can be maintained during the stretching process.12,15 The
principle of the strain gauge sensor is based on increase in
Acoustic Based Methods
the resistance of a conductor when the area of the con-
Respiratory sound can be measured using a microphone ductor is increased during the respiration process.
placed either close to the respiratory airways or over the Normally the inspiratory thoracic and abdominal ex-
throat to detect the variation of sound. Then a frequency pansion is almost synchronous. However, if the upper
analysis and estimation of the loudness of the sound can be airway is partially obstructed, there may be a change in
carried out.8 the phase angle and timing of the movements of the thorax
Werthammer et al.9 reported a respiratory sounds and abdomen.16 The movements become asynchronous,
measurement system to detect sleep apnea in infants. that is, the thorax moves inwards, and the abdomen out-
The system depended on recording a signal derived from wards. During expiration this pattern is then reversed.
breathing sounds from the nose. This method was applied Thoraco-abdominal asynchrony is a normal finding in
to eight premature infants. Snorting, speaking, crying, infants in whom chest wall compliance is greater17 and
coughing, etc. had a negative effect on the operation of is exacerbated by respiratory disease or respiratory muscle
the system. weakness.18
Corbishley and Rodriguez-Villegas10 proposed a mini- Nepal et al.19 studied the abdominal strain gauge trans-
aturized and wearable respiration monitoring respiration ducer for measuring respiration rate. The strain gauge was
system. It used a microphone mounted on the neck to strapped around the patient’s chest and changes in thoracic
obtain the largest breathing acoustic. or abdominal circumference during breathing were
measured. The method involved a classification algorithm
Airflow Based Methods to separate respiratory signals as apnoea, respiration, or
respiration with motion, by using a second order autore-
Airflow can be detected because exhaled air is warmer,
gressive modeling and zero cross algorithm.
has higher humidity and contains more CO2 than inhaled
air. These variations can be used for indicating the respir-
Transcutaneous CO2 Monitoring
atory rate. Most airflow-sensing methods need a sensor,
attached to the airways.7 The measurement of the airflow In transcutaneous CO2 monitoring a heated electrode
can be achieved by using a nasal or oronasal thermistor (about 428C) is applied to the skin (usually an arm). This
which detects changes in temperature between the inhaled method relies on the diffusion of gas to the skin and
and exhaled air. This gives a semi-quantitative estimate of provides an overall estimate of change in CO2 level rather
Pediatric Pulmonology
 Respiration Rate Monitoring Methods 525
26
than minute by minute readings. The electrode is sur- calculation of the mean electrical axis variation. The
rounded by a solution to provide conductivity. Care needs study concluded that single lead respiration rate estimates
to be taken to avoid skin burning on sensitive and neonatal are more robust than the methods based on the mean
skin.12 Transcutaneous CO2 monitoring therefore allows electrical axis.
measurements of consequences of abnormal ventilation Tarassenko et al.27 looked at three methods: The EDR,
rather than a measure of the respiratory rate itself. the electrical impedance pneumography (IP) signal across
the chest, and the ECG or the changes in light absorption
Oximetry Probe (SpO2) Based which is known as photoplethysmogram (PPG) across a
finger. They obtained estimates of the breathing rate by
Blood oxygen saturation (SpO2) measurement is
adding the individual estimated outputs for both the IP and
another technique for monitoring the consequences of
PPG channels after applying the Kalman filters for both
abnormal ventilation. When air enters the lungs its oxygen
waveforms. A limitation of these methods is that move-
binds to the hemoglobin in red blood cells. The oxygen is
ment artefacts introduce errors in estimating the breathing
then transported throughout the body in arterial blood. A
rates.
pulse oximeter uses the red and infrared frequencies to
determine the percentage of hemoglobin in the blood that
is saturated with oxygen. This percentage is called blood NONCONTACT RESPIRATORY MONITORING
saturation, or SpO2.20 An oximeter simultaneously METHODS
displays the SpO2 level as well as the pulse rate and
Radar Based Respiration Rate Monitoring
plethysmogram.
There have been studies indicating that respiration rate Greneker28 reported one of the first noncontact respir-
can be extracted from plethysmograms. Plethysmograms ation rate monitoring systems. The system was called the
from ten healthy adults were processed using wavelet Radar Vital Signs Monitor (RVSM). It was developed to
transforms.21 Respiration waveform was observed in monitor the performance of Olympic athletes at distances
the plots of the wavelet transforms. In another study exceeding 10 m. The RVSM detected breathing–induced
involving 14 infants, of median age 2 days, the feasibility movements of the chest using the Doppler phenomenon. A
of extracting respiratory information from the plethsmo- limitation of this method was motion artefact which cor-
gram traces was also demonstrated.22 The magnitude rupted the breathing signals. There have been no published
frequency spectra of the plethsmogram traces showed studies describing the use of this method in children.
peaks associated with respiration rate.
Optical Based Respiration Rate Monitoring
Electrocardiogram (ECG) Derived Respiration Rate
Aoki et al.29 reported a non-restrictive visual sensing
This method is based on the fact that respiration has a method to detect the respiration pattern by using a fiber
modulating effect on the ECG. In this respiration rate grating (FG) vision sensor and processor unit. Their sys-
monitoring approach, ECG electrodes are attached to tem consisted of two parts. The first was the FG projecting
the subject in order to record an ECG. By measuring device. This provided an array of invisible infra-red light
the fluctuation in ECG, the respiration rate can be derived. spots (wavelength 810 nm). The second part was a
This technique is called ECG-Derived Respiration (EDR) Charge-Coupled Device (CCD) camera with an optical
and is based on a process known as sinus arrhythmia, that band-pass filter. Infrared light was used to project a set of
is, the modulation of ECG by the breathing process.23 bright spots on the subject, while the CCD camera was
EDR is believed to be based on small ECG morphology used to capture the scene of bright spots. The moving
changes during the respiratory cycle caused by movement distances of bright spots in each image were extracted and
of the heart position relative to the electrodes and the analyzed to monitor respiration.
change in lung volume. Principal component analysis Nakajima et al.30 used a static camera to detect thoracic
has been used to identify which ECG lead was most movements to determine respiration rate. The projection
effective before extracting the respiration rate.24 of the surface of the thorax was represented as a region
EDR monitoring has also been performed by using a with a range of brightness intensities. Respiration was
single-channel that did not have to be a precordial lead.25 monitored by quantifying the variations of the locations
In contrast to a number of other studies that used ECG of the image intensities over time.
characteristic waves (e.g., QRS complex), this study used Non-invasive optical methods have not previously been
the higher order statistics of ECG recording (such as the described in children, however, in a study carried out by
4th order cumulant). our group, a high speed desktop computer connected to a
A study carried out a quantitative comparison of EDR video camera (webcam) was used to record respiration
monitoring techniques based on direct measurement of the related chest and abdominal movements in children in a
modulation components versus techniques based on sleep unit.31 The video consisted of a series of sequential
Pediatric Pulmonology
526 AL-Khalidi et al.

images, each marked with an individual time stamp (t),

corresponding to the time of the recording. An algorithm
was designed to subtract the current image at time t (i.e.,
imgt) and the image a few time frames before at time
t td (i.e., imgt td). The value of td was determined
practically. An increase in the value of td caused faster
algorithm operation, but decreased the resolution of the
algorithm in detecting movements. In this study a value of
0.5 sec was used for td as it was observed to provide an
acceptable compromise between the resolution and speed. Fig. 2. Plot x(t) against time during three respiration cycles.
The resulting difference images (imgdiff) were then thresh-
olded to produce binary images (imgx). The sum of pixel
values in each binary image was determined to produce a placed on the mask so that breathing could be detected
set of data values (x). The magnitude of each data value in when the child’s head turned. The problem with this
x represented the movement between the time t and t td. method was that a mask had to be placed close to the
These values were then plotted against time to obtain a child’s face.
respiration signal. Zhu et al.33 developed an infrared imaging based respir-
A plot of the values of x during exhalation and inhala- ation rate monitoring method. They designed a tracking
tion is provided in Figure 1. The vertical axis of the plot algorithm that could follow facial features related to
represents chest and abdominal movements. Two lobes respiration. These features were selected manually from
were observed, the lobe representing inhalation being a reference image (i.e., the first image in the video) by
larger. During exhalation, the chest’s wall initially moves specifying three windows. Two of these windows covered
slowly inward, its movement increases with time, reaching the areas between the bridge of nose and the inner corner of
a peak and then, the amount of movement decreases. A the eyes (i.e., the periorbital regions) and represented the
similar process occurs during inhalation, but this time the warmest facial areas. Another window was placed on the
chest’s wall moves outward. Figure 2 shows the plot of x(t) apex of nose to represent the coolest facial area. Their
during three respiration cycles. An algorithm was devel- algorithm tracked these three windows in the following
oped to extract the respiration rate from the recorded recorded images. The respiration signal was obtained from
signal in real-time. a rectangular region under the nose.
Instead of using a focal plane array of mid-wave infra-
Thermal Sensor and Thermal Imaging Based red sensors, Chekmenev et al.34 used a thermal camera
Respiration Rate Monitoring consisting of a focal plane array for a long-wave infra-red
(6–15 mm) sensor. They measured the temperature
Hsu and Chow32 have reported a thermal sensor based
changes around the neck region, carotid vessel complex,
respiration rate monitoring system. In this approach there
and the nasal region. The selection of these regions was
was no contact with the child’s skin. The sensor could
carried out manually. A wavelet analysis technique was
detect temperature changes induced by respiration and
developed to extract the ECG and the respiration rate.
then the data were corrected and analyzed simultaneously
In a study carried out by our group, the FLIR A40
by a personal computer that was linked to a central nursery
thermal camera was used to monitor respiration related
room. To avoid missing the breathing signals, an ellipsoid-
skin surface temperature variation in area centered on the
shaped mask was made and the thermo sensors were
tip of the nose.35 The thermal sensitivity of the camera is
0.08 K. The camera was fixed on a tripod in front of the
subject at a distance of about 1 m. The image capture rate
was 50 frames/sec, thus providing 3,000 images/min. The
recording time was 2 min. The images were segmented
and then an algorithm was used to locate and track a
circular area centered on the tip of the nose. The segmen-
tation and tracking algorithms are described Al-Khalidi
et al.36 The chosen area was divided into eight equal
concentric segments as shown in Figure 3. The pixel
values within each of the eight segments were averaged
to obtain a single value representing the skin temperature
in that segment. The process was repeated for each image.
Plots of the average temperature against time for the
Fig. 1. A plot of values of x during inhalation and exhalation. segments were obtained (see Fig. 4). These plots
Pediatric Pulmonology
 Respiration Rate Monitoring Methods 527

the clearest respiration signal. An algorithm was produced

to automatically extract the respiration rate from the
recorded signal.

DISCUSSION
Respiration rate monitoring has evolved from the days
when it could only be measured by placing the back of the
hand close to the nose to monitor exhaled air. It is very
important to develop a reliable, accurate and objective
method of monitoring respiratory rate in children which is
non-invasive and therefore will not cause distress to the
child which may lead to inaccuracies in the measurement.
Fig. 3. (a) A thermal image with tip of the nose represented by a The gold standard is still to place a thermistor in a nostril to
circle, (b) the eight segments of the selected respiration region. monitor exhaled air. However, a variety of contact and
noncontact respiration monitoring devices have been
developed.
represented the respiration signal associated with the seg- Several techniques are available for measurement of
ments. The respiration signal reduces in amplitude during respiratory rate via nasal prongs, masks, thermistor, or
inhalation and increases in amplitude during exhalation. respiratory impedance plethysmography. These methods
The clarity of the signals varied, with segments 3 and 7 all involve some contact with the child. A pneumotacho-
providing the most clear signals and segment 1 providing graph has the advantage that it gives quantitative assess-
the least clear signal. This result indicated that for respir- ments of flows, from which respiratory rate can be derived,
ation monitoring, it would be advisable to consider all the and may be important in a research setting. The thermistor
eight segments and then to select the segments that provide is a commonly used technique to detect temperature

Fig. 4. Respiration signals obtained from segments 1 to 8 (from top to bottom respectively).

Pediatric Pulmonology
528 AL-Khalidi et al.

changes breath by breath at the nostril, thereby giving the Studies are still ongoing to produce more effective respir-
number of breaths per minute. This technique has been ation monitoring devices.
formally validated in the measurement of respiratory rate
for the purposes of sleep studies.11 A semi-quantitative REFERENCES
measure of airflow and tidal volume can be derived from
1. Lausted CG, Jonston AT. Respiratory system. In: Bronzino JD,
respiratory inductance plethysmography (RIP). The tech- editor. Biomedical engineering fundamentals. CRC Press; 2006
nique has been demonstrated to work well in measuring Chapter 7, 7.1–7.17.
tidal volume in infants.37 2. Fieselmann JF, Hendryx MS, Helms CM, et al. Respiratory rate
However, from a clinical perspective the main disad- predicts cardiopulmonary arrest for internal medicine patients. J
vantage of these methods is that they require connection to Gen Intern Med 1993;8:354–360.
3. Subbe CP, Davies RG, Williams E, et al. Effect of introducing
the child, some to the facial area, which disturbs many the Modified Early Warning score on clinical outcomes, cardio-
children and may be poorly tolerated. In addition, in pulmonary arrests and intensive care utilisation in acute medical
infants and children, the added dead space of the equip- admissions. Anaesthesia 2003;58:797–802.
ment required for some of these techniques may have an 4. Goldhill DR, McNarry AF, Manersloot G, et al. A physiologically-
influence on breathing patterns.38 based early warning score for ward patients: the association
between score and outcome. Anaesthesia 2005;60:547–553.
There are extensive opportunities to further improve 5. Cretikos M, Chen J, Hillman K, et al. The objective medical
noncontact respiration monitoring schemes. Radar, infra- emergency team activation criteria: a case–control study. Resus-
red imaging and optical imaging have all been used to citation 2007;73:62–72.
monitor respiration rate. Advances in computing technol- 6. Murthy R, Pavlidis I, Tsiamyrtzis P. Touchless monitoring of
ogy have resulted in realization of complex algorithms for breathing function. Engineering in Medicine and Biology Society,
2004. IEMBS apos;04. 26th Annual International Conference of
respiration monitoring. For example, it has been possible the IEEE. Vol. 2, No. 1, pp. 1196–1199.
to use 3-D vision algorithms and two web cameras to 7. Folke M, Cernerud L, Ekstro M, Hok B. Critical review of non-
monitor respiration in manikins.36 invasive respiratory monitoring in medical care. Med Biol Eng
Currently noncontact respiration monitoring methods Comput 2003;41:377–383.
have not yet reached the level of maturity that can be used 8. Shneerson MJ. Sleep medicine: a guide to sleep and its
disorders, handbook. 2nd edition. UK: Blackwell Publishing
routinely in clinical environments. Concerns related to Ltd; 2005.
patient safety, electromagnetic interference with existing 9. Werthammer J, Krasner J, DiBenedetto J, Stark RA. Apnea
medical equipments, and complexity of use have been monitoring by acoustic detection of airflow. Pediatrics 1983;71:
factors in their slow take up in clinical environments. 53–55.
With further development, noncontact methods will 10. Corbishley P, Rodriguez-Villegas E. Breathing detection: toward a
miniaturised, wearable, battery-operated monitoring system. IEEE
gradually become more and more viable and will Trans Biomed Eng 2008;55:196, 204.
effectively complement contact based respiration rate 11. Storck K, Karlsson M, Ask P, Loyed D. Heat transfer evaluation
monitoring methods. of the nasal thermistor technique. IEEE Trans Biomed Eng
We have developed two novel methods for noncontact 1996;43:1187–1191.
respiratory rate monitoring in children which, whilst cur- 12. Lee-Chiong LT, editor. Sleep: a comprehensive handbook. John
Wiley and Sons Inc.; 2006.
rently are sophisticated, high technology devices are 13. Tobin MJ. Respiratory monitoring in the intensive care unit. Am
undergoing further evaluation in acute clinical settings Rev Repatriation Dis 1988;138:1625–1642.
and will serve as a basis for future development to produce 14. Folke M, Granstedt F, Hok B, Scheer H. Comparative provocation
an inexpensive, user friendly device for use in the clinical test of respiratory monitoring methods. J Clin Monit Comput
environment. 2002;17:97–103.
15. Blom AJ. Monitoring of respiration and circulation 2004. CRC
Press LLC; Boca Raton, London, New York, Washington, DC.
CONCLUSIONS 16. Prisk GK, Hammer J, Newth CJ. Techniques for measurement of
thoracoabdominal asynchrony. Pediatr Pulmonol 2002;34:462–
Monitoring respiration rate is an important task when 472.
evaluating a subject’s health. Respiration rate monitoring 17. Mayer OH, Clayton RG, , Jawad AF, McDonough JM, Allen JL.
devices can be classified by a number of ways depending Respiratory inductance plethysmography in healthy 3- to 5-year-
old children. Chest 2003;124:1812–1819.
on the manner of their use and their operation. In this 18. Stromberg NO, Nelson N. Thoracoabdominal asynchrony in
review article, they were grouped into contact and non- small children with lung disease—methodological aspects and
contact. This review has highlighted the advances made to the relationship to lung mechanics. Clin Physiol 1998;18:447–456.
improve the effectiveness of respiration monitoring. The 19. Nepal K, Biegeleisen E, Ning T. Apnea Detection and Respiration
potential for noncontact respiration monitoring is empha- rate Estimation through Parametric Modeling. In: Proceedings of
the 28th IEEE Annual Northeast Bioengineering Conference,
sized. Noncontact respiration rate monitoring devices Philadelphia, Pennsylvania; 2002. pp. 277–278.
have a distinct advantage over contact methods, especially 20. Konica Minolta Sensing, Inc., Basic understanding of the pulse
in children, as they cause least disturbance to the patient. oximeter.

Pediatric Pulmonology
 Respiration Rate Monitoring Methods 529

21. Leonard P, Beattie TF, Addison PS, Watson JN. Standard pulse 31. Tan KS, Saatchi R, Elphick H, Burke D. Real-time vision based
oximeters can be used to monitor respiratory rate. Emerg Med J respiration monitoring system. IEEE, IET International
2003;20:524–525. Symposium on Communication Systems, Networks and Digital
22. Wertheim D, Olden C, Savage E, Seddon P. Extracting respiratory Signal Processing 2010; 846-850 ISBN-Hardcopy: 988-1-86135-
data from pulse oximeter plethysmogram traces in newborn 370-2.
infants. Arch Dis Child Fetal Neonatal Ed 2009;94:301–303. 32. Hsu CH, Chow CJ. Design and clinic monitoring of monitoring of a
23. Moody G, Mark R, Bump M, Weinstein J, Berman A, Mietus J, newly developed non-attached infant apnea monitor. Centre for
Goldberger A. Clinical validation of ECG-derived respiration Biomedical Engineering, Taiwan, Province De Chine 2005;17:
(EDR) technique. Comput Cardiol 1986;13:507–510. 126–133.
24. Mazzanti B, Lamberti C, de Bie J. Validation of an ECG-derived 33. Zhu Z, Fei J, Pavlidis I. Tracking Human Breath in Infrared
respiration monitoring method. Comput Cardiol 2003;30:613– Imaging. In: Proceeding of the 5th IEEE, Symposium on Bioin-
616. formatics and Bioengineering (BIBE’05). Organization IEEE
25. Ding S, Zhu X, Chen W, Wei D. Derivation of respiratory signal Computer Society Washington, DC, USA 2005;19:227–231.
from single-channel ECGs based on source Ding statistics. Int J 34. Chekmenev Y, Rara H, Farag A. Non-contact, wavelet-based
Bioelectromagn 6:43–49. measurement of vital signs using thermal imaging. ICGST Int J
26. O’Brien C, Heneghan C. Comparison of algorithms for estimation Graph Vision Image Process 2006;6:25–30.
of respiratory signal from the surface electrocardiogram. Comput 35. Saatchi R, AL-Khalidi F, Burke D, Elphick H. Thermal Image
Biol Med 2007;37:305–314. Analysis of the Skin Surface Centred on the Tip of the Nose for
27. Tarassenko L, Mason L, Townsend N. Multi-sensor fusion for Respiration Monitoring. ICEDSP International Conference on
robust computation of breathing rate. IEEE Soc 2002;38:1314– Computer Systems and Applications, Manipal, India; 2009.
1316. 36. Al-Khalidi FQ, Saatchi R, Burke D, Elphick H. Facial tracking
28. Greneker EF. Radar sensing of heartbeat and respiration at a method for noncontact respiration rate monitoring, IEEE, IET
distance with applications of the technology. Radar 1997;97: International Symposium on Communication Systems, Networks
150, 154. and Digital Signal Processing, Newcastle, UK. 21–23 July, 2010.
29. Aoki H, Takemura Y, Mimura K, Nakajima M. Development of pp. 827–830. ISBN-Hardcopy: 978-1-86135-370-2.
non-restrictive sensing system for sleeping person using fibre 37. Hershenson MB, Colin AA, Wohl ME, Stark AR. Changes in the
grating vision sensor. In: Proceedings of 2001 International contribution of the rib cage to tidal breathing during infancy. Am
Symposium on Micromechatronics and Human Science, Nagoya, Rev Respir Dis 1990;141:922–925.
Japan: 155: 160. 38. Brooks LJ, DiFiore JM, Martin RJ. Assessment of tidal volume
30. Nakajima K, Matsumoto Y, Tamura T. Development of real time over time in preterm infants using respiratory inductance plethys-
image sequence analysis for evaluating posture change and respir- mography, The CHIME Study Group. Collaborative Home Infant
atory rate of a subject in bed. Physiol Meas 2001;22:21–28. Monitoring Evaluation. Pediatr Pulmonol 1997;23:429–433.

Pediatric Pulmonology