pulse oxymeters and does not bring any alteration of sleep, as racic and abdominal wall expansion (by respiratory inductive
there is no variation of pressure at the fingertip. Catchside et al. plethysmography), oronasal airflow (by nasal pressure), arterial
have shown that PWA has better correlation with acoustical- oxygen saturation SaO2 (by pulse oximetry) and PPG (sampling
induced arousals undetected by EEG [12]. In patients with SDB, frequency = 75 Hz with a resolution of 16 bits/sample). Both
it has been reported that respiratory events and arousals induce red and infrared (IR) PPG signals were collected during PSG
more changes in PWA than that in the heart rate (HR) [13]. recording; however, we have used only the IR channel in this
The peak-to-peak time interval (PPI) measured from a PPG study. All subjects were free of any cardiac history. Diagnosis
signal can be used for extracting the instantaneous pulse rate was based on clinical symptoms and PSG outcomes. Arousals
variability (PRV) signal and has been reported as a surrogate were scored manually according to ASDA criteria using the
for ECG-based heart rate variability (HRV) analysis in healthy EEG signal and marked using Somnologica software. In total,
subjects [13]. Recently, Lazro et al. have reported that use of 2917 arousal events were found across all ten subjects of this
PRV improves the accuracy of obstructive sleep apnea subject study.
classification with respect to HRV in children [14]. Therefore,
the correlation of PPI with HRV, which has a clinical relevance B. Schematic Diagram
in sleep studies, indicates that it could be used to detect RERAs.
The schematic diagram to detect arousal using features of
In addition to PWA and PPI features, in this study we also ana-
the PPG signal is shown in Fig. 2. At first, the features are
lyzed the performance of Area (area under PWA curve) feature
extracted from the PPG signal and an arousal model is developed
in respiratory arousal detection. Although no specific physiolog-
based on an individual feature. After detecting events, closer
ical relation of Area has been reported yet, we hypothesize that
events (multiple events within 10 s) are merged together as
it could be an important feature in detecting respiratory arousal,
a postprocessing step. To obtain the arousal detection using
since it contains information of the morphology of PPG cycle.
multiple features, detected arousal events of individual features
The purpose of this study is to investigate the ability of var-
are combined. Finally, the performance (accuracy, false positive
ious PPG features to detect the respiratory arousals as a new
rate, true positive rate, and missed positive rate) of the model
technique for the screening of sleep quality or fragmentation.
is measured based on the original (EEG-based score) arousal
Examples of respiratory events and arousal in a polysomonog-
events.
raphy (PSG) recording are shown in Fig. 1.
1) PPG Features: In this study, three different time domain
features are extracted from the PPG signal. The example of the
II. DATA AND METHODS PPG signal and the extracted features from the PPG signal are
shown in Fig. 3. A brief description of the features is given
A. Subjects and Data
below.
In total, ten polysomnographic recordings were used in this a) Peak-to-peak interval (PPI): The PPI is defined as the time
study. PSGs were collected from ten patients with sleep dis- difference between two consecutive peaks of the PPG signal.
ordered breathing at Charite Hospital, Berlin, Germany. The At first, the peak of each cycle of the PPG signal (peakAmp)
research protocol was approved by Charite Hospital Ethics in was detected and the time stamps of all peakAmp points were
Human Research Committee. Each PSG study included elec- stored in a vector. The PPI was calculated as the time differ-
troencephalogram (channel C3-A2 and C4-A1), left and right ence between consecutive peakAmp points (see Fig. 3). In order
electro-oculogram (EOG), leg movements, body positions, tho- to reduce the error in calculating the PPI, the PPG signal was
KARMAKAR et al.: DETECTION OF RESPIRATORY AROUSALS USING PHOTOPLETHYSMOGRAPHY (PPG) SIGNAL IN SLEEP APNEA PATIENTS 1067
III. RESULTS
The Somnologica snapshot of output signals at different steps
of arousal detection using PWA feature is shown in Fig. 6. The
panel representing EEG (C4) signal represents the reference
arousal events. The PWA feature extracted from the IR signal
is shown as Raw PWA, which was resampled into a 1-Hz
signal and shown as PWA (1 Hz). The rectangular box in panel
PWA (1 Hz) represents the 25 s window used to calculate the
PCparam and the output is shown in panel %Change PWA.
The straight line in %Change PWA represents the PCthresh
value and was used to detect the arousal events, which is shown
in Raw Arousal Events PWA. Finally, Arousal Events PWA
was generated by merging multiple raw arousal events within a
10 s time period and this was the detected arousal event used to
analyze in the next step.
Briefly, the arousal detection using PPI, PWA, Area, and the
combination of PWA and Area is shown in Fig. 7. Steps in-
Fig. 5. Steps for calculating PCp a ra m (i). Top panel shows the PWAre f based volving the calculation of features and resampling into a 1-Hz
on which PC is calculated which is shown in the bottom panel. The bottom panel
also shows the average of percentage change PC1av g and PC2 av g and their signal, which is shown in Fig. 6, are not repeated in Fig. 7. Pan-
difference as PCp a ra m (i). els Raw Events (PPI), Raw Events (PWA), and Raw Events
(Area) were calculated using a threshold value on %Change
PPI, %Change PWA, and %Change Area data, respec-
and the second average (PC2avg ) was calculated as the average tively, which are shown in detail in Fig. 6 for the PWA feature.
of feature value from the 15th to 20th seconds. Finally, the Arousal Events PPI, Arousal Events PWA, and Arousal
PCparam for that window (PCparam (i)) was calculated as the Events Area were calculated by merging multiple events within
difference between PC1avg and PC2avg . Since the calculation 10 s of Raw Events (PPI), Raw Events (PWA), and Raw
of PCparam at any time requires at least 9 s prior and 15 s Events (Area), respectively. Merging of Raw Events (Area)
postfeatures, the first 9 and last 15 values of P Cparam signals is shown using a solid circle and an arrow in Fig. 7. Arousal
were set to zero, and therefore, it could not be used to detect Events (PWA + Area) was calculated by combining events
arousal for that period. After calculating the PCparam signal, detected in Arousal Events PWA and Arousal Events Area,
the arousal events were detected using a predefined percentage i.e., any event in Arousal Events (PWA + Area) was presented
parameter threshold value (PCthresh ). Any PCparam value larger only when the event was detected by both PWA and Area fea-
than PCthresh was detected as the raw arousal events. The raw tures. This is shown in Fig. 7 using the dotted circle and an
arousal events were then merged into a single event, if multiple arrow.
KARMAKAR et al.: DETECTION OF RESPIRATORY AROUSALS USING PHOTOPLETHYSMOGRAPHY (PPG) SIGNAL IN SLEEP APNEA PATIENTS 1069
Fig. 6. Output signals at different steps of arousal detection using the PWA feature.
Fig. 7. Arousal events detection using PPI, PWA, and Area features. Arousal events detection by filtering raw arousal events detected using PPI, PWA, and Area
features.
1070 IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, VOL. 18, NO. 3, MAY 2014
Fig. 8. Accuracy and false rate of arousal detection for all ten subjects using (a) PPI, (b) Area, (c) PWA, and (d) PWA + Area.
The performance of PWA, Area, PPI, and PWA + Area fea- model showed higher accuracy than the false rate, which was
tures (accuracy and false detection rate) with varying threshold also opposite to the pattern found with PWA- and Area-based
of PC is shown in Fig. 9. Using the PWA feature [see Fig. 9(a)], models. As a result, with increasing PCthresh value, the false
maximum accuracy obtained was around 93% with a false rate rate becomes higher than accuracy for PPI model; however,
of around 89%. The false rate drops below 49% at 80% accuracy both values become very low, which indicates that the PPI-
level with PCthresh = 30%. Moreover, for the PWA feature, the based model underestimates or misses most of the events for
drop in accuracy is much slower compared to the drop in the any PCthresh value. Using both PWA and Area features, the
false rate with increasing PCthresh value. This indicates that the maximum accuracy obtained is around 92% with a false rate of
use of a relatively higher threshold value will reduce the false around 85%. The false rate drops below 47% at an 80% accuracy
alarm while using PWA. level with a threshold change of 28%.
The Area feature showed a maximum accuracy value of PPG-based arousal indexes agreed well across a wide range
around 93% with a false rate of around 85%. The false rate drops of decision thresholds (ROC = 0.91), and the decision threshold
below 50% at an 80% accuracy level with PCthresh = 32%. Sim- for the final analyses (PCthresh = 25%) was chosen with a sensi-
ilar to the PWA feature, accuracy of Area feature also decayed tivity of 68.1% and a specificity of 95.2%. At PCthresh = 25%,
slowly compared to the false rate. However, decreasing the rate the results showed an accuracy of 14.68%, 85.15%, 86.93%,
of false rate for PWA feature with increasing PCthresh value was and 50.79% with a false rate of 21.80%, 55.41%, 64.78%, and
higher than Area feature (see Fig. 8). 50.79% for PPI, PWA, Area, and PWA + Area features, respec-
Using the PPI feature, the maximum accuracy obtained was tively. The actual arousal index and number of arousals for
around 88% with a false rate of around 86%. The false rate individual subjects are compared to that of the arousal detection
drops below 75% at an 80% accuracy level with PCthresh = 7%. model developed in this study and summarized in Table I.
In contrast to the PWA and Area features, the accuracy drops The PPI feature severely underestimated the number of
faster than the false rate for PPI feature with increasing PCthresh arousal events except for subject 1, which was highly over-
value. In addition, at the lowest PCthresh value, the PPI-based estimated (344 events detected against 117 actual events). This
KARMAKAR et al.: DETECTION OF RESPIRATORY AROUSALS USING PHOTOPLETHYSMOGRAPHY (PPG) SIGNAL IN SLEEP APNEA PATIENTS 1071
TABLE I
ACTUAL AND CALCULATED NUMBER OF AROUSALS AND AROUSAL INDEX OF ALL 10 SUBJECTS. COMB MEANS BOTH
PWA AND AREA FEATURES ARE USED TO DETECT THE AROUSAL
Actual Calculated
Subject Sleep duration
No of Arousal No of Arousal Arousal Index
ID (hours:mins)
Arousal Index PPI/PWA/Area/Comb PPI/PWA/Area/Comb
1 7:14 117 16.18 344/140/320/114 47.47/19.32/44.16/15.73
2 6:51 362 52.85 203/445/472/427 29.62/64.94/68.88/62.31
3 7:03 459 65.11 51/509/499/496 7.22/72.07/70.65/70.23
4 10:54 464 42.57 160/582/680/527 14.68/53.38/62.37/48.34
5 6:48 368 54.12 20/443/453/423 2.94/65.02/66.49/62.09
6 7:04 217 30.71 63/461/464/429 8.91/65.20/65.63/60.68
7 7:25 243 32.76 74/489/453/430 9.96/65.82/60.97/57.88
8 6:49 172 25.23 40/421/402/386 5.87/61.74/58.95/56.61
9 8:24 412 49.05 65/326/427/305 7.74/38.81/50.83/36.31
10 6:19 303 47.97 87/358/365/323 13.74/56.53/57.64/51.00
indicates that the PPI threshold PCthresh = 25% resulted in very The PWA feature is a well-known autonomic marker and
low accuracy with relatively higher false rate, which was shown has been previously used for the detection of respiratory events
in Fig. 8(a). In contrast, the Area feature overestimated the num- and arousals in patients with SAS [18]. However, in that study
ber of arousal events for all subjects. For example, the maximum PWA was used as an additional parameter along with airflow
overestimation happens for subject number 6, with 464 events and oxygen desaturation to score respiratory events. The PPI
detected for 217 actual events (see Table I). This means that it feature was previously used as a surrogate of HRV analysis [13]
maintains a high accuracy with relatively higher false rate for and recently reported to classify obstructive sleep apnoea in
PCthresh = 25% [see Fig. 8(b)]. The PWA feature also showed children [14]. However, the PRV feature used in that study is
overestimation except for subject 9, where it detected only 326 different from the PPI feature used in this study and no result has
events compared to 412 actual events. However, the overesti- been reported in detecting respiratory events or arousal in the
mation of the PWA feature is less compared to the Area feature. sleep disordered adult population. Moreover, use of another PPG
This resulted in a similar accuracy value with lower false rate feature (Area) in arousal detection is reported for the first time
than the Area feature at PCthresh = 25% for the PWA feature in this study to our knowledge. We are convinced that our new
[see Fig. 8(c)]. Interestingly, the combination of PWA and Area analysis allows a better quantification of respiratory arousals.
features reduces the false rate, while keeping accuracy at the Previous works have suggested that there could be other mark-
same level [see Fig. 8(d)]. This means that the use of the com- ers of arousal such as PTT, peripheral arterial tonometry, etc.,
bination of these two features increases the reliability of the which are much easier to measure than EEG and which could as-
detection of arousal events. As a result, although PWA + Area sess the true extent of sleep fragmentation [19][22]. However,
(Comb) overestimates the number of arousal events for most of in those studies PTT signal was visually analyzed for respiratory
the subjects (except subjects 1 and 9), the number of estimated event or arousal detection. In this study, we used PPG features
events were closest to the actual event number for all subjects which evaluate changes in peripheral arterial volume derived
(shown in a bold format in Table I). However, the combination features as markers for automatic respiratory arousal detection.
of all three features decreases accuracy and increases the false On the other hand, PAT has been used with other markers for ex-
rate due to the influence of PPI feature [see Fig. 8(a)] and the ample oximetry and actigraphy (Watch-PAT) in order to detect
results are not incorporated in this study. The arousal index pre- sleep apnea and arousal [22].
sented in Table I was calculated as the number of events/hour The direct relation between the drop in PWA values and
and it showed the same pattern as the number of arousal events respiratory arousals was previously reported in several stud-
discussed previously. ies [18], [23], [24]. In addition to the PWA feature, we have
found a similar drop in PPI and Area values. However, the main
problem in arousal event detection using PPG features lies in the
IV. DISCUSSION selection of a threshold value. An overestimated threshold value
The PPG-based algorithm was designed to differentiate be- reduces both the false rate and accuracy of the model. Although
tween EEG arousal and non-EEG arousal conditions. This study reduction in false rate is desirable for any model, the accuracy
demonstrated that it is possible to detect visually scored EEG reduction is not. In contrast, the underestimated threshold value
arousals with an algorithm based on the PPG signal alone. Al- resulted in increasing both the accuracy and false rate, which is
though close association of cortical and autonomic arousals also undesirable. Therefore, in this study we have calculated the
has been repeatedly shown in the past [16], [17], some distinct accuracy and false rate over a range of threshold value for all
events may occur independently (i.e., cortical arousal without PPG features [see Fig. 8(a)(d)]. From the characteristic curve
autonomic arousal and vice versa). At least some of the disagree- of accuracy and false rate, we have found that it is not possible
ment between visual EEG arousal scoring and PPG-based scor- to define a fixed threshold value for all features. In this study,
ing may be attributed to these independently occurring events. we have selected the threshold value for which the model shows
1072 IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, VOL. 18, NO. 3, MAY 2014
at least 80% accuracy with a minimum false rate and we have Finally, the PPG-based respiratory arousal detection model is
found 30%, 32%, 7%, and 28% PCthresh values for PPG features a simple and could be a promising alternative to conventional
PWA, Area, PPI, and PWA + Area respectively. electroencephalogram (EEG)-based arousal detection systems
In order to get better detection results, the following recom- useful for monitoring sleep health at home and ambulatory
mendations in selecting thresholds are proposed. These param- conditions.
eters are as follows.
1) Percentage change parameter (PCparam ): For calculating
PCparam , the window length can be varied. In this study, REFERENCES
we have varied the window length from 15 to 20 s for the [1] American Academy of Sleep Medicine Taskforce, Sleep-related breath-
PPI feature and compared their performance. From those ing disorders in adults: Recommendations for syndrome definition and
measurement techniques in clinical research, Sleep, vol. 22, pp. 667
results, we found that window length of 20 s performed 689, 1999.
better than other window lengths. [2] J. M. Shineerson, Sleep Medicine: A Guide to Sleep and its Disorder.
2) Percentage change threshold (PCthresh ): In this study, Oxford, U.K.: Blackwell, 2005, pp. 230231.
[3] E. A. Phillipson and C. E. Sullivan, Arousal: The forgotten response to
we have used PCthresh = 25%, for all features. This may respiratory stimuli, Amer. Rev. Respir. Dis., vol. 118, pp. 807809, 1978.
however be changed to achieve the desired accuracy or [4] J. E. Remmers, W. J. deGroot, E. K. Sauerland, and A. M. Anch, Patho-
false rate values which may be verified with other clinical genesis of upper airway occlusion during sleep, J. Appl. Physiol., vol. 44,
pp. 931938, 1978.
measures of daytime sleepiness. Also, some false positives [5] C. Guilleminault and M. Rosekind, The arousal threshold: Sleep depri-
could be autonomic arousals which need to be verified in vation, sleep fragmentation, and obstructive sleep apnea syndrome, Bull
future. Eur. Physiopathol. Respir., vol. 17, pp. 341349, 1981.
[6] L. C. Johnson and A. Lubin, The orienting reflex during waking and
In this study, we have used three PPG features (PPI, PWA, sleeping, Electroencephalogr Clin. Neurophysiol., vol. 22, pp. 1121,
and Area) for arousal detection [(see Fig. 8(a)(c)]. PWA and 1967.
Area have shown better performance (higher accuracy and lower [7] N. J. Douglas and S. E. Martin, Arousals and the sleep apnea/hypopnea
syndrome, Sleep, vol. 19, no. 10, pp. S196S197, 1996.
false rate) compared to the PPI feature at PCthresh = 25%. Af- [8] S. E. Martin, P. K. Wraith, I. J. Deary, and N. J. Douglas, The effect of
ter investigating the possible combinations of those features, we nonvisible sleep fragmentation on daytime function, Amer. J. Respir. Crit.
have come up with a conclusion that combining PWA and Area Care Med., vol. 155, pp. 15961601, 1997.
[9] N. A. Collop, Portable monitoring for the diagnosis of obstructive sleep
provides better accuracy with lower false rate in arousal detec- apnea, Curr. Opin. Pulm. Med., vol. 14, pp. 525529, 2008.
tion [see Fig. 8(d)]. From the subjective analysis of the number [10] D. Pitson, N. Chhina, S. Knijn, M. van Herwaaden, and J. Stradling,
of arousals and arousal index, it is obvious that the developed Changes in pulse transit time and pulse rate as markers of arousal from
sleep in normal subjects, Clin. Sci., vol. 87, pp. 269273, 1994.
model overestimates the actual arousal number and index except [11] P. G. Catcheside, S. C. Chiong, J. Mercer, N. A. Saunders, and
subjects 1 and 9 (see Table I). This is in line with the findings R. D. McEvoy, Noninvasive cardiovascular markers of acoustically in-
reported in previous study that changes in PWA were also seen duced arousal from non-rapid-eye-movement sleep, Sleep, vol. 25, no. 7,
pp. 797804, 2002.
in the absence of conventionally scored EEG arousals which [12] J. Haba-Rubio, G. Darbellay, F. R. Herrmann, J. G. Frey, A. Fernandes,
can be attributed to the subcortical arousal [24], [25]. J. M. Vesin, J. P. Thiran, and J. M. Tschopp, Obstructive sleep apnea
One limitation of this study is the small number of subjects syndrome: Effect of respiratory events and arousal on pulse wave ampli-
tude measured by photoplethysmography in NREM sleep, Sleep Breath,
used to validate the PPG-based respiratory arousal detection vol. 9, no. 2, pp. 7381, 2005.
model developed. However, most of the subjects had moderate [13] M. Nitzan, S. Turivnenko, A. Milston, A. Babchenko, and Y. Mahler,
to high respiratory arousal index; therefore, the number of events Low-frequency variability in the blood volume and in the blood volume
pulse measured by photoplethysmography, J. Biomed. Opt., vol. 1, no. 2,
(2917) analyzed in this study was large. In addition, the visual pp. 223229, 1996.
scoring of arousal depends on the scorer [26]. It was our aim to [14] J. Lazaro, E. Gil, J. M. Vergara, and P. Laguna, OSAS detection in chil-
have just one visual scorer for this validation study. However, dren by using PPG amplitude fluctuation decreases and pulse rate vari-
ability, Comput. Cardiol., vol. 39, pp. 185188, 2012.
further validation of the developed model needs to be done with [15] A. Zacharia, J. Haba-Rubio, R. Simon, G. John, P. Jordan, A. Fernandes,
more subjects and variation in the number of arousal per subject J. M. Gaspoz, J. G. Frey, and J. M. Tschopp, Sleep apnea syndrome: Im-
(very small arousal index to very large arousal index) to get a proved detection of respiratory events and cortical arousals using oxymetry
pulse wave amplitude during polysomnography, Sleep Breath., vol. 12,
better understanding of the correlation between PPG features no. 1, pp. 3338, 2008.
and sleep disturbances, which would include a comparison of [16] A. Delessert, F. Espa, A. Rossetti, G. Lavigne, M. Tafti, and R. Heinzer,
different scorers on scoring the arousal [26]. Pulse wave amplitude drops during sleep are reliable surrogate markers
of changes in cortical activity, Sleep, vol. 33, no. 12, pp. 16871692,
In conclusion, the PWA and Area features-based model for 2010.
arousal detection showed better performance compared to the [17] J. Haba-Rubio, G. Darbellay, F. R. Herrmann, J. G. Frey, A. Fernandes,
PPI-based model. Combining the PWA and Area features in- J. M. Vesin, J. P. Thiran, and J. M. Tschopp, Obstructive sleep apnea
syndrome: Effect of respiratory events and arousal on pulse wave ampli-
creased the reliability of the model, i.e., a reduced false detec- tude measured by photoplethysmography in NREM sleep, Sleep Breath,
tion rate without much effect on the accuracy. This was due to vol. 9, no. 2, pp. 7381, 2005.
the overestimating nature of both PWA- and Area-based mod- [18] D. J. Pitson, A. Sandell, R. van den Hout, and J. R. Stradling, Use of pulse
transit time as a measure of inspiratory effort in patients with obstructive
els. Although PPG-based respiratory arousal detection models sleep apnoea, Eur. Respir. J., vol. 8, pp. 16691674, 1995.
mostly overestimate the number of actual events, which can [19] J. Argod, J. L. Pepin, and P. Levy, Differentiating obstructive and central
be attributed to sub-cortical arousal, future exploration of sub- sleep respiratory events through pulse transit time, Amer. J. Respir. Crit.
Care Med., vol. 158, pp. 17781783, 1998.
cortical and cortical arousal can provide explicit explanation [20] J. Argod, J. L. Pepin, R. P. Smith, and P. Levy, Comparison of esophageal
of the reason behind the overestimation found in this study. pressure with pulse transit time as a measure of respiratory effort for
KARMAKAR et al.: DETECTION OF RESPIRATORY AROUSALS USING PHOTOPLETHYSMOGRAPHY (PPG) SIGNAL IN SLEEP APNEA PATIENTS 1073
scoring obstructive nonapneic respiratory events, Amer. J. Respir. Crit. Thomas Penzel (M92SM06) received the Gradu-
Care Med., vol. 162, pp. 8793, 2000. ation degree from physics, human biology, and phys-
[21] T. Penzel, K. Kesper, I. Pinnow, H. F. Becker, and C. Vogelmeier, Periph- iology from the University Marburg, Marburg, Ger-
eral arterial tonometry, oximetry and actigraphy for ambulatory recording many, in 1986, 1991, and 1995, respectively.
of sleep apnea, Physiol. Meas., vol. 25, pp. 10251036, 2004. He has been with the University of Marburg since
[22] S. M. Hilton, The defence-arousal system and its relevance for circulatory 1984 at the Department of Physics, Physiology, and
and respiratory control, J. Exp. Biol., vol. 100, pp. 159174, 1982. the sleep laboratory until 2006. Since 2006 he has
[23] E. Sforza, F. Chapotot, S. Lavoie, F. Roche, R. Pigeau, and A. Buguet, been the Director of Research of the interdisciplinary
Heart rate activation during spontaneous arousals from sleep: Effect of sleep medicine center, Charite University Hospital,
sleep deprivation, Clin. Neurophysiol., vol. 11, pp. 244251, 2004. Berlin, Germany. He is the Editor-in-Chief of the Ger-
[24] P. G. Catcheside, S. C. Chiong, J. Mercer, N. A. Saunders, and man journal Somnologie and on the editorial board of
R. D. McEvoy, Noninvasive cardiovascular markers of acoustically other journals. His research interests include biomedical signals, biomedical
induced arousal from non-rapid-eye-movement sleep, Sleep, vol. 7, devices, the cardiovascular, and the neural system, all related to sleep.
pp. 797804, 2002. Dr. Penzel received the Bial award for clinical medicine in Portugal, in 2001,
[25] J. Hedner, H. Ejnell, J. Sellgren, T. Hedner, and G. Wallin, Is high and the Bill Gruen Award for Innovations in Sleep Research by the American Sleep
fluctuating muscle nerve sympathetic activity in the sleep apnoea syn- Research Society, in 2008, and the Somnus Award in Germany, in 2012. He
drome of pathogenetic importance for the development of hypertension? holds several functions in sleep medicine societies, Chair of the ESRS Sleep
J. Hypertens Suppl., vol. 6, pp. S529S531, 1988. Medicine Committee, Secretary of the German Sleep Society, and others.
[26] J. S. Loredo, J. L. Clausen, S. Ancoli-Israel, and J. E. Dimsdale, Night-
to-night arousal variability and interscorer reliability of arousal measure-
ments, Sleep, vol. 22, no. 7, pp. 916920, 1999.
Chandan Karmakar (S07M11) received the Christoph Schobel received the Medical degree
at ChariteUniversitaetsmedizin Berlin, Berlin,
B.Sc. Eng. degree in computer science and engineer-
Germany, in 2008.
ing from the Shah Jalal University of Science and
He is currently doing internship in internal
Technology, Sylhet, Bangladesh, in 1999, and the
Doctor of Engineering degree from The University medicine at the Department for Cardiology + An-
giology, Interdisciplinary Center for Sleep Medicine,
of Melbourne, Parkville, Vic., Australia, in 2012.
Charite Universitaetsmedizin Berlin and working
He is currently working as a Postdoctoral Research
in EU-funded projects (DAPHNet, HIVE).
Fellow in the Department of Electrical and Elec-
tronic Engineering, The University of Melbourne,
Australia. He has published one book and more than
40 peer-reviewed journal and conference articles in
the research field of physiological signal processing and modeling. His research
interests include biomedical devices and signal processing, cardiovascular and
neural system related to sleep disordered breathing, human gait dysfunctions,
cardiovascular diseases, and diabetic autonomic neuropathy.
Marimuthu Palaniswami (S84M87SM94
F12) received the M.E. degree from the Indian In-
stitute of Science, Bangalore, India, the M.Eng.Sc.
degree from The University of Melbourne, Parkville,
Ahsan Khandoker (M07SM12) received the Vic., Australia, and the Ph.D. degree from the Uni-
B.Sc. degree in electrical and electronic engineering versity of Newcastle, Callaghan, N.S.W., Australia.
from the Bangladesh University of Engineering and He is currently a Professor with the Department
Technology (BUET), Dhaka, Bangladesh, in 1996, of Electrical and Electronic Engineering, The Univer-
the M. Eng.Sc. from Multimedia University (MMU), sity of Melbourne. He has published more than 400
Malaysia, in 1999, and the M. Eng. and Doctor of En- refereed research papers and leads one of the largest
gineering degree in physiological engineering from funded Australian Research Council, Research Net-
the Muroran Institute of Technology, Muroran, Japan, work on Intelligent Sensors, Sensor Networks, and Information Processing
in 2004. programme. His research interests include support vector machines sensors and
He is currently an Assistant Professor in the De- sensor networks, IoT, machine learning, neural network, pattern recognition,
partment of Biomedical Engineering, Khalifa Uni- signal processing, and control.
versity, Abu Dhabi, UAE. He is also a Senior Research Fellow for Australian Dr. Palaniswami has been a Grants Panel Member for The National Science
Research Council Research Networks on Intelligent Sensors, Sensor Networks, Foundation, an Advisory Board Member for the European FP6 grant center, a
and Information Processing (ISSNIP), the University of Melbourne, Parkville, Steering Committee Member for National Collaborative Research Infrastructure
Vic., Australia. He has published 35 peer-reviewed journal articles and more Strategy, Great Barrier Reef Ocean Observing System, Smart Environmental
than 75 conference papers the research field of physiological signal processing Monitoring and Analysis Technologies, and a Board Member for Informa-
and modeling in sleep disordered breathing, diabetic autonomic neuropathy, tion Technology and supervisory control and data acquisition companies. He
fetal cardiac disorders and human gait dysfunction, and is passionate about re- has been funded by several ARC and industry grants (over 40 m) to conduct
search helping clinicians to noninvasively diagnose diseases at early stage. He research in sensor network, Internet of things (IoT), health, environmental, ma-
was also with several Australian Medical device manufacturing industries, as chine learning, and control areas. He is representing Australia as a core partner
well as hospitals as a Research Consultant focusing on integration of technology in European Union FP7 projects such as SENSEI, SmartSantander, Internet of
in clinical settings. Things Initiative, and SocIoTal.
The author has requested enhancement of the downloaded file. All in-text references underlined in blue are linked to publications on ResearchGate.