<!--#set var="TITLE" value="Biometric Human Identification based on ECG"-->
<!--#include virtual="/head.shtml"-->
<p align="center"><b>T. Lugovaya</b><br>
Department of Applied Mathematics and Computer Science,
Electrotechnical University "LETI", Saint-Petersburg, Russian Federation</p>
<div class="notice">
<p>
This material originally appeared in
<blockquote>
Lugovaya T. Biometric human identification based on ECG. [Master's thesis]
Dept of Applied Mathematics and Computer Science, Electrotechnical University
"LETI", St Petersburg, Russian Federation; 2005.
</blockquote>
<p>
Please cite this publication when referencing this material.
</div>
<p>
<a name="abstract">
<h2>Abstract</h2></a>
<p><i>This research investigates the feasibility of using the electrocardiogram
(ECG) as a new biometric for human identification. It is well known that the
shapes of the ECG waveforms depend on human heart anatomic features and are
different for different persons. But it is unclear whether such differences
can be used to identify different individuals. This research demonstrates that
it is possible to identify a specific person in a predetermined group using a
one-lead ECG. A one-lead ECG is a one-dimensional, low-frequency signal that
can be recorded from electrodes on the hands. ECG fragments containing QRS
complexes, P and T waves extracted from the ECG are processed by principal
component analysis and classified using linear discriminant analysis. Using
this method on a predetermined group of 90 subjects, the experimental results
showed that the rate of correct identification was 96%.</i></p>
<a name="introduction">
<h2>Introduction</h2></a>
<p>Biometric technologies are among fast-developing fields of information
security, gradually entering into all spheres of human activity. Today only
three biometric methods have proved their efficiency, namely, identification
based on fingerprints, iris or retina, and face. Hand geometry, voice, writing
and typing dynamics, etc. are also useful, depending on the purpose and range
of application.</p>
<p>This research aims to develop an identification system based on ECG (figure
1). ECG is assumed as an almost unique human characteristic because morphology
and amplitudes of registered cardiac complexes are governed by multiple
individual factors, in particular, by formation and position of the heart,
presence and nature of pathologies, etc.</p>
<center>
<img src="images/ecg.png" alt="[example of ECG]">
<p><b>Figure 1.</b> Example of ECG with standard notations.</p>
</center>
<a name="system">
<h2>Identification system</h2></a>
<p> The system feasibility was discussed in [1, 2]. This study suggests other
data interpretation and classification techniques, with the system tested on a
higher level of live input data. The respective findings are compared below
(table 2).</p>
<p> The identification system uses a classical scheme including data
preprocessing, formation of input data space, transition to reduced feature
space, ECG cycle classification and ECG record identification.</p>
<p> For usability, it is necessary to be able to collect the ECG easily and
quickly. The data collected for this study comprise the <a href="./">ECG-ID
Database</a>, consisting of 320 single-lead ECG recordings from 90 subjects,
each 20 seconds long, sampled at 500 Hz with 12-bit precision. Since
single-lead ECGs vary significantly within an individual depending on the lead
(the locations of the electrodes used to observe the ECG), the choice of lead
is important. Lead I is the potential difference between the left and right
hands (LA - RA). It was chosen because it is easily measured and it is not
sensitive to minor variations in electrode locations.</p>
<a name="preprocessing">
<h3>Data preprocessing</h3></a>
<p> The raw ECG is often rather noisy and contains distortions of various
origins. Both high and low frequency noise components are present (figure
3).</p>
<center>
<table>
<tr>
<td align=center><img src="images/ecg_noise_net.png" alt="[power-line noise]"><br>A. ECG with power-line noise</td>
<td align=center><img src="images/ecg_noise_high.png" alt="[high-frequency noise]"><br>B. ECG with high-frequency noise</td>
</tr>
<tr>
<td align=center><img src="images/ecg_noise_both.png" alt="[power-line and high-frequency noise]"><br>C. ECG with both power-line and high-frequency noise</td>
<td align=center><img src="images/ecg_noise_drift.png" alt="[isoline drift]"><br>D. ECG with baseline drift</td>
</tr>
</table>
<b>Figure 2.</b> Examples of noisy ECGs.
</center>
<p>Frequency-selective signal filtering was implemented using a set of adaptive bandstop and low-pass filters (figure 3).</p>
<center>
<table>
<tr>
<td><img src="images/noise_filter_1.png" alt="[signal filtering 1]"></td>
<td><img src="images/noise_filter_2.png" alt="[signal filtering 2]"></td>
</tr>
</table>
<b>Figure 3.</b> Examples of frequency-selective signal filtering results.
</center>
<p> Baseline drift correction was implemented using multilevel one-dimensional
wavelet analysis. The original signal was decomposed at level 9 using
biorthogonal wavelets. The signal reconstructed using final approximation
coefficients is assumed to be the drifting baseline, which is subtracted from
the original signal (figure 4). This method shows good results in both cases of
clear and rather noisy ECG signals.</p>
<center>
<table>
<tr>
<td><img src="images/wavelet_drift_correction.png" alt="[wavelet drift correction]"></td>
<td><img src="images/wavelet_drift_correction_noisy.png" alt="[wavelet drift correction noisy]"></td>
</tr>
</table>
<b>Figure 4.</b> Examples of baseline drift correction results.
</center>
<a name="feature_space">
<h3>Initial feature space</h3></a>
<p>This section focuses on initial feature space formation. Obviously
information on cardiac performance is basically held in the cardiac cycle
fragment containing the QRS complex and P- and T-waves (referred to here as the
PQRST-fragment). Therefore the process begins with extraction of a set of
R-peak synchronized PQRST-fragments (figure 5). The PQRST-fragment length is
fixed at 0.5 sec (250 samples).</p>
<center>
<table>
<tr>
<td><img src="images/pqrst_set_1.png" alt="[PQRST-fragment extraction]"></td>
<td><img src="images/pqrst_set_2.png" alt="[PQRST-fragment extraction]"></td>
</tr>
</table>
<b>Figure 5.</b> R-peak detection and PQRST-fragment extraction.
</center>
<p>PQRST fragments are used as informative features. Therefore extracted PQRST
fragments are processed to enhance their similarity as follows:
<table>
<tr>
<td>1.</td>
<td colspan=3>Correcting PQRST fragment mutual "vertical" shift due to residual
baseline drift:</td>
</tr>
<tr>
<td></td>
<td colspan=3 align="center"><img src="images/pqrst_set_vertical_correction.png"
alt="[vertical shift correction]"></td>
</tr>
<tr>
<td>2.</td>
<td colspan=3>Culling distorted (due to breathing or motion artifacts) and
pathological PQRST-fragments:</td>
</tr>
<tr>
<td></td>
<td colspan=3 align="center"><img src="images/pqrst_set_atypical.png"
alt="[atypical PQRST-fragments]"></td>
</tr>
<tr>
<td>3.</td>
<td colspan=3>Correcting PQRST-fragments depending on heart rate using Bazett's formula:</td>
</tr>
<tr>
<td></td>
<td align="center"><img src="images/pqrst_set_rate_original.png" alt="[heart rate influence]"></td>
<td align="center"><img src="images/arrow_hor.png"></td>
<td align="center"><img src="images/pqrst_set_rate_corrected.png" alt="[heart rate correction]"></td>
</tr>
</table>
</p>
<p>Thus in the initial feature space (dimension N=250) the ECG appears as a set
of PQRST-fragments with each seen as a separate pattern at subsequent system
stages, to be interpreted and classified independently.</p>
<a name="reduction">
<h3>Feature space reduction</h3></a>
<p>The feature space is reduced using Principal Component Analysis (PCA) so
that its dimension can be reduced to 30 (with the Kaiser criterion) or even
to 10 (with the scree test).</p>
<p>Alternately, the feature space can be reduced with a wavelet transform (WT)
providing the same space reduction but with slightly poorer final
identification.</p>
<a name="classification">
<h3>Classification and Identification</h3></a>
<p>The resulting PQRST-fragment patterns are then classified in reduced feature
space using linear discriminant analysis.</p>
<p>At the final stage, the ECG record identification is based on PQRST-fragment
classification results.</p>
<a name="results">
<h2>Results</h2></a>
<p>Experimental studies involved 90 human volunteers. ECG records were made in
the sitting position. Heart rate and physical and emotional state were not
limited. The collected data set contains 320 records, of which 200 records
were assigned to the training set and 120 records to the test set.
Differentiation between the training and test sets aimed to provide for maximum
performance complexity, i.e., maximum difference between records in different
sets both in monitoring time and human physical state.</p>
<p>Averaged results of series of experiments on PQRST-fragment classification
and ECG record identification with different feature space reduction methods
are tabulated in Table 1. As shown there, the best results were obtained using
30 principal components, yielding 96% correct ECG identification in the test
set.</p>
<center>
<table border="1" cellspacing="2" bordercolor="#808080" cellpadding="3">
<tbody>
<tr>
<td align="center" rowspan="2">Reduction technique</td>
<td align="center" rowspan="2">Number of features</td>
<td align="center" colspan="2">PQRST-fragments classification, %</td>
<td align="center">ECG identification, %</td>
</tr>
<tr>
<td align="center">Training set</td>
<td align="center">Test set</td>
<td align="center">Test set</td>
</tr>
<tr>
<td align="center">PCA</td>
<td align="center">10</td>
<td align="center">99</td>
<td align="center">85</td>
<td align="center">89</td>
</tr>
<tr>
<td align="center">WT</td>
<td align="center">9</td>
<td align="center">98</td>
<td align="center">79</td>
<td align="center">82</td>
</tr>
<tr>
<td align="center"><b>PCA</b></td>
<td align="center"><b>30</b></td>
<td align="center">99</td>
<td align="center">91</td>
<td align="center"><b>96</b></td>
</tr>
<tr>
<td align="center">WT</td>
<td align="center">34</td>
<td align="center">99</td>
<td align="center">88</td>
<td align="center">91</td>
</tr>
</tbody></table>
<b>Table 1. </b>Experimental results.
</center></p>
<p>Additionally, results from [1, 2] and this research are compared in Table 2.
<center>
<table border="1" cellspacing="2" bordercolor="#808080" cellpadding="3">
<tbody>
<tr>
<td align="center">Research</td>
<td align="center">Class count</td>
<td align="center">Identification results, %</td>
</tr>
<tr>
<td align="center">[1]</td>
<td align="center">20</td>
<td align="center">98</td>
</tr>
<tr>
<td align="center">[2]</td>
<td align="center">9</td>
<td align="center">95</td>
</tr>
<tr>
<td align="center">this</td>
<td align="center">90</td>
<td align="center">96</td>
</tr>
</tbody></table>
<b>Table 2. </b> Results comparison.
</center></p>
<p> As a result of this research a recognition system was developed to
solve the problem of biometric human identification based on ECG on a
sufficiently large set of input data. The findings represent primary
arguments for using ECG as a biometric characteristic in various
biometric access control problems, opening up a brand new perspective
for the study of biometric technologies, with potential applications in
security and modern amenity systems.</p>
<a name="refs">
<h2>References</h2></a>
<p>[1] Biel L, Pettersson O, Philipson L, Wide P. ECG analysis: a new approach
in human identification. IEEE Transactions on Instrumentation and Measurement
2001 June; 50(3):808-812.
<p>[2] Yi WJ, Park KS, Jeong DU. Personal identification from ECG measured
without body surface electrodes using probabilistic neural networks. Proc
2003 World Congress on Medical Physics and Biomedical Engineering,
Sydney, Australia, 2003 August. </p>