16-09-2016, 11:19 AM
Wavelet Based ECG Steganography for Protecting
Patient Confidential Information in Point-of-Care
Systems
1454835978-43.WaveletBasedECGSteganographyforProtecting.pdf (Size: 541.49 KB / Downloads: 20)
Abstract—With the growing number of aging population and
a significant portion of that suffering from cardiac diseases, it
is conceivable that remote ECG patient monitoring systems are
expected to be widely used as Point-of-Care (PoC) applications
in hospitals around the world. Therefore, huge amount of ECG
signal collected by Body Sensor Networks (BSNs) from remote
patients at homes will be transmitted along with other physiological
readings such as blood pressure, temperature, glucose level
etc. and diagnosed by those remote patient monitoring systems.
It is utterly important that patient confidentiality is protected
while data is being transmitted over the public network as well
as when they are stored in hospital servers used by remote
monitoring systems. In this paper, a wavelet based steganography
technique has been introduced which combines encryption and
scrambling technique to protect patient confidential data. The
proposed method allows ECG signal to hide its corresponding
patient confidential data and other physiological information
thus guaranteeing the integration between ECG and the rest. To
evaluate the effectiveness of the proposed technique on the ECG
signal, two distortion measurement metrics have been used: the
Percentage Residual Difference (PRD) and the Wavelet Weighted
PRD (WWPRD). It is found that the proposed technique provides
high security protection for patients data with low (less
than 1% ) distortion and ECG data remains diagnosable after
watermarking (i.e. hiding patient confidential data) and as well
as after watermarks (i.e. hidden data) are removed from the
watermarked data.
INTRODUCTION
T
HE number of elderly patients is increasing dramatically
due to the recent medical advancements. Accordingly, to
reduce the medical labor cost, the use of remote healthcare
monitoring systems and Point-of-Care (PoC) technologies
have become popular [1], [2]. Monitoring patients at their
home can drastically reduce the increasing traffic at hospitals
and medical centres. Moreover, Point-of-Care solutions can
provide more reliability in emergency services as patient
medical information (ex. for diagnosis) can be sent immediately
to doctors and response or appropriate action can be
taken without delay. However, Remote health care systems
are used in large geographical areas essentially for monitoring
purposes, and, the Internet represents the main communication channel used to exchange information. Typically, patient biological
signals and other physiological readings are collected
using body sensors. Next, the collected signals are sent to
the patient PDA device for further processing or diagnoses.
Finally, the signals and patient confidential information as
well as diagnoses report or any urgent alerts are sent to the
central hospital servers via the Internet. Doctors can check
those biomedical signals and possibly make a decision in case
of an emergency from anywhere using any device[3].
Using Internet as main communication channel introduces
new security and privacy threats as well as data integration
issues. According to the Health Insurance Portability and
Accountability Act (HIPAA), information sent through the
Internet should be protected and secured. HIPAA mandates
that while transmitting information through the internet a
patient’s privacy and confidentiality be protected as follows:
[4]:
1) Patient privacy: It is of crucial importance that a patient
can control who will use his/her confidential health
information, such as name, address, telephone number,
and Medicare number. As a result, the security protocol
should provide further control on who can access
patient’s data and who cannot.
2) Security: The methods of computer software should
guarantee the security of the information inside the
communication channels as well as the information
stored on the hospital server.
Accordingly, it is of crucial importance to implement a
security protocol which will have powerful communication
and storage security [5].
Several researchers have proposed various security protocols
to secure patient confidential information. Techniques used
can be categorized into two subcategories. Firstly, there are
techniques that are based on encryption and cryptographic
algorithms. These techniques are used to secure data during
the communication and storage. As a result, the final data
will be stored in encrypted format [4], [6], [2], [7]. The
disadvantage of using encryption based techniques is its large
computational overhead. Therefore, encryption based methods
are not suitable in resource-constrained mobile environment.
Alternatively, many security techniques are based on hiding
its sensitive information inside another insensitive host data
without incurring any increase in the host data size and
huge computational overhead. These techniques are called
steganography techniques. Steganography is the art of hiding secret information inside another type of data called host
data [8]. However, steganography techniques alone will not
solve the authentication problem and cannot give the patients
the required ability to control who can access their personal
information as stated by HIPAA.
In this paper, a new security technique is proposed to
guarantee secure transmission of patient confidential information
combined with patient physiological readings from
body sensors. The proposed technique is a hybrid between
the two preceding categories. Firstly, it is based on using
steganography techniques to hide patient confidential information
inside patient biomedical signal. Moreover, the proposed
technique uses encryption based model to allow only the
authorized persons to extract the hidden data. In this paper,
the patient ECG signal is used as the host signal that will
carry the patient secret information as well as other readings
from other sensors such as temperature, glucose, position,
and blood pressure. The ElectroCardioGram (ECG) signal is
used here due to the fact that most of the healthcare systems
will collect ECG information. Moreover, the size of the ECG
signal is large compared to the size of other information.
Therefore, it will be suitable to be a host for other small
size secret information. As a result, the proposed technique
will follow HIPAA guidelines in providing open access for
patients biomedical signal but prevents unauthorized access to
patient confidential information.
In this model body sensor nodes will be used to collect
ECG signal, glucose reading, temperature, position and blood
pressure, the sensors will send their readings to patient’s PDA
device via Bluetooth. Then , inside the patient’s PDA device
the steganography technique will be applied and patient secret
information and physiological readings will be embedded
inside the ECG host signal. Finally, the watermarked ECG
signal is sent to the hospital server via the Internet. As a
result, the real size of the transmitted data is the size of the
ECG signal only without adding any overhead, because the
other information are hidden inside the ECG signal without
increasing its size. At hospital server the ECG signal and its
hidden information will be stored. Any doctor can see the watermarked
ECG signal and only authorized doctors and certain
administrative personnel can extract the secret information and
have access to the confidential patient information as well as
other readings stored in the host ECG signal. This system is
shown in Fig 1.
The proposed steganography technique has been designed in
such a way that guarantees minimum acceptable distortion in
the ECG signal, Furthermore, it will provide the highest security
that can be achieved. The use of this technique will slightly
affect the quality of ECG signal. However, watermarked ECG
signal can still be used for diagnoses purposes as it is proven
in this paper. In this work the following research questions are
answered:
• Can the proposed technique protect patient confidential
data as explained in the HIPAA security and privacy
guidelines?
• What will be the effect on the original ECG signal after
applying the proposed steganography technique in terms
of size and quality?
Rest of the paper is organized as follows. Section II briefly
discusses the related works and what other researchers did in
this area. In section III we discuss the basic system design,
the embedding process (i.e patient sensitive data into ECG
signal) and the extraction process. Next, in section IV security
analysis is proposed. Then Section V explains diagnosability
measurement. Section VI shows the results of PRD calculated
before and after secret data extraction. Finally, section VII
concludes the paper.
II. RELATED WORK
There are many approaches to secure patient sensitive data
[9], [7], [2], [10]. However, one approach [11], [12], [13]
proposed to secure data is based on using steganography
techniques to hide secret information inside medical images.
The challenging factors of these techniques are how much
information can be stored, and to what extent the method is
secure. Finally, what will be the resultant distortion on the
original medical image or signal.
Kai-mei Zheng and Xu Qian [13] proposed a new reversible
data hiding technique based on wavelet transform. Their
method is based on applying B-spline wavelet transform on the
original ECG signal to detect QRS complex. After detecting
R waves, Haar lifting wavelet transform is applied again on
the original ECG signal. Next, the non QRS high frequency
wavelet coefficients are selected by comparing and applying
index subscript mapping. Then, the selected coefficients are
shifted one bit to the left and the watermark is embedded.
Finally, the ECG signal is reconstructed by applying reverse
haar lifting wavelet transform. Moreover, before they embed
the watermark, Arnold transform is applied for watermark
scrambling. This method has low capacity since it is shifting
one bit. As a result only one bit can be stored for each ECG
sample value. Furthermore, the security in this algorithm relies
on the algorithm itself, it does not use a user defined key.
Finally, this algorithm is based on normal ECG signal in which
QRS complex can be detected. However, for abnormal signal
in which QRS complex cannot be detected, the algorithm will
not perform well.
H. Golpira and H. Danyali [12] proposed a reversible blind
watermarking for medical images based on wavelet histogram
shifting. In this work medical images such as MRI is used as
host signal. A two dimensional wavelet transform is applied
to the image. Then, the histogram of the high frequency subbands
is determined. Next, two thresholds are selected, the
first is in the beginning and the other is in the last portion
of the histogram. For each threshold a zero point is created
by shifting the left histogram part of the first threshold to
the left, and shifting the right histogram part of the second
threshold to the right. The locations of the thresholds and the
zero points are used for inserting the binary watermark data.
This algorithm performs well for MRI images but not for ECG
host signals. Moreover, the capacity of this algorithms is low.
Moreover, no encryption key is involved in its watermarking
process.