Академический Документы
Профессиональный Документы
Культура Документы
Alessandro Bac`a, Giorgio Biagetti, Marta Camilletti, Paolo Crippa, Laura Falaschetti,
Simone Orcioni, Luca Rossini, Dario Tonelli, Claudio Turchetti
PPG Channels Accel. Channels This approximation is equivalent to assume the signals are in
the subspace
hSi = span (s1 . . . sP ) , (4)
Windowing generated by the basis S = [s1 . . . sP ] RN P where
s1 . . . sP are the most significative components of the motion
signal. hSi represents the SMS.
Filtering
g =m+e , g RN (5)
SVD SVD SVD
where e is the heart-rate signal, m the artifact and g the
FFT
PPG signal. As m belongs to the subspace hSi, then the
FFT FFT
Peak Peak Peak corresponding Hankel data matrix G = g (1) . . . g (L) ,
Finder Finder Finder G RN L can be written as
G = SA + E (6)
Motion Artifact
with A = a(1) . . . a(L) RP L , E = e(1) . . . e(L)
Removal
RN L .
It is worth to note that, as S can be decomposed in two
Tracker orthonormal subsets S, S
S = [s1 , . . . , sP , sP +1 , . . . sN ] = [S S ] (7)
BPM
with S = [sP +1 , . . . sN ], E decomposes into two compo-
Fig. 1: Flow chart of CARMA algorithm. nents, belonging to the two orthogonal subspaces hSi, hS i
B
E = [S S ] = SB + S B (8)
2.1. Subspace of motion signals B
Given the accelerometer signals x, y, z the first step is to where B, B are the corresponding coordinates, so that G is
determine the corresponding subspace hSi they belong to, given by
that S that generates hSi.
(1)is a basis To this end let X =
x . . . x(L) , Y = y (1) . . . y (L) , Z = z (1) . . . z (L) be G = SA + SB + S B = S(A + B) + S B . (9)
the Hankel data matrices of the three signals respectively,
where x(i) , y (i) , z (i) RN , i = 1, . . . , L, L > N represent Assuming the component SB of E belonging to the sub-
the observations achieved from the three-axes accelerometer. space hSi is negligible when comparing with the artifact com-
The complete matrix of sample signals ponent SA we have
where S = [s1 . . . sN ], R = [r1 . . . rN ] and the eigenvalues 2.3. Removing the closest subspace to hSi from SVD de-
i are in decreasing order 1 2 . . . N . Since, in composition of G
general, it results P +1 , . . . , N < with 0 we can write
It is well known that the SVD decomposes the given space
P RN in orthonormal subspaces. To this end let
H
X
= i si riT . (3)
i=1
G = U V T (12)
2697
23rd European Signal Processing Conference (EUSIPCO)
2698
23rd European Signal Processing Conference (EUSIPCO)
250
Table 1: Average performance on recordings 112 [BPM].
200
Method MAE RMSE
CARMA 2.26 3.63
TROIKA 2.34 3.07
Hz]
150
frequency [60
REFERENCES
0
0 50 100 150 200 250 300
time [s]
[1] B. S. Kim and S. K. Yoo, Motion artifact reduction
in photoplethysmography using independent component
Fig. 2: AT algorithm at work with different starting points (red
analysis, IEEE Trans. Biomedical Engineering, vol. 53,
lines). Regardless of initialization, it always converges to the
no. 3, pp. 566568, March 2006.
same track (green line), which is very close to the ground truth
(black line). [2] J. Y. A. Foo, Comparison of wavelet transformation
and adaptive filtering in restoring artefact-induced time-
related measurement, Biomedical Signal Processing
3. EXPERIMENTAL RESULTS and Control, vol. 1, no. 1, pp. 9398, 2006.
[3] P. T. Gibbs, L. B. Wood, and H. H. Asada, Active
To validate the effectiveness of the algorithm we tested it on motion artifact cancellation for wearable health monitor-
the 12-subject dataset used in [9], and on an extra track (sub- ing sensors using collocated MEMS accelerometers, in
ject 13) in which the subject did a different kind of exercise. Smart Structures and Materials. International Society for
Figure 3 shows examples of the tracking performed. The Optics and Photonics, 2005, vol. 5765, pp. 811819.
red and amber stars represent the frequency estimates com-
ing from the peak finders for the two PPG channels. The blue [4] B. Lee et al., Improved elimination of motion artifacts
circles the frequency estimate of the most prominent peak in from a photoplethysmographic signal using a Kalman
the accelerometer channels (actually, P = 10 peaks are used smoother with simultaneous accelerometry, Physiolog-
in the artifact removal, but only the first is shown to avoid ical Measurement, vol. 31, no. 12, pp. 1585, 2010.
cluttering the figures). The green solid line is our HR esti- [5] M. Raghuram, K. V. Madhav, E. H. Krishna, and K. A.
mate. As a reference, the ECG-derived true HR is the black Reddy, Evaluation of wavelets for reduction of motion
solid line. The tracking is generally good, and even if the al- artifacts in photoplethysmographic signals, in 10th Int.
gorithm is sometimes driven off the correct track (because the Conf. Information Sciences Signal Processing and their
MAR algorithm was overly aggressive and removed a valid Applications (ISSPA), May 2010, pp. 460463.
HR candidate, or the MA frequency lay so close to the HR [6] M. Raghuram, K. Sivani, and K. A. Reddy, E2MD
that the SVD frequency resolution couldnt resolve them), it for reduction of motion artifacts from photoplethysmo-
soon recovers. graphic signals, in Int. Conf. Electronics and Commu-
A comparison of its performance with those of [9] are re- nication Systems (ICECS), Feb 2014, pp. 16.
ported in Tab. 1, which reports the mean absolute error (MAE)
[7] M. Raghuram et al., HHT based signal decomposi-
and the root mean square error (RMSE), and in more detail for
tion for reduction of motion artifacts in photoplethysmo-
each subject in Tab. 2. Despite being much simpler and not
graphic signals, in IEEE Int. Instrum. Meas. Tech. Conf.
requiring an initialization phase, our algorithm attained com-
(I2MTC), May 2012, pp. 17301734.
parable scores, often remarkably better.
[8] M. R. Ram et al., A novel approach for motion arti-
fact reduction in PPG signals based on AS-LMS adap-
4. CONCLUSIONS tive filter, IEEE Trans. Instrum. Meas., vol. 61, no. 5,
pp. 14451457, May 2012.
A novel algorithm for HR estimation from PPG signals has
been presented. It removes motion artifacts and adaptively [9] Z. Zhang, Z. Pi, and B. Liu, TROIKA: A general frame-
tracks HR exploiting dominant frequencies of SVD decom- work for heart rate monitoring using wrist-type photo-
position to separate the PPG in two subspaces, overcoming plethysmographic signals during intensive physical exer-
many limitations of already existing techniques. Experimen- cise, IEEE Trans. Biomedical Engineering, vol. 62, no.
tal evaluations performed on datasets recorded from subjects 2, pp. 522531, Feb 2015.
2699
23rd European Signal Processing Conference (EUSIPCO)
Method Subj1 Subj2 Subj3 Subj4 Subj5 Subj6 Subj7 Subj8 Subj9 Subj10 Subj11 Subj12
CARMA 2.58 1.48 1.40 2.47 1.54 3.24 1.01 1.19 0.93 6.28 1.68 3.30
TROIKA 2.29 2.19 2.00 2.15 2.01 2.76 1.67 1.93 1.86 4.70 1.72 2.84
subject 6 subject 7
200 200
180 180
160 160
140 140
Hz]
Hz]
120 120
frequency [60
frequency [60
100 100
80 80
60 60
40 40
20 20
0 0
0 50 100 150 200 250 300 0 50 100 150 200 250 300
time [s] time [s]
subject 9 subject 10
200 200
180 180
160 160
140 140
Hz]
Hz]
120 120
frequency [60
frequency [60
100 100
80 80
60 60
40 40
20 20
0 0
0 50 100 150 200 250 300 0 50 100 150 200 250 300
time [s] time [s]
subject 12 subject 13
200 200
180 180
160 160
140 140
Hz]
Hz]
120 120
frequency [60
frequency [60
100 100
80 80
60 60
40 40
20 20
0 0
0 50 100 150 200 250 300 0 50 100 150 200 250 300
time [s] time [s]
2700