Printer Friendly

A Survey On Medical Image Protection Using Various Steganography Techniques.

INTRODUCTION

Communicating the digital information becomes fast by frequent access capability. In modern communication the security threats must be resisted during end-end protection. Data hiding and cryptography are the two techniques which are used to secure the communication. In cryptography, the plain data is changed into an unreadable form is called cipher data. The demerit of cryptography, the third party is conscious about the communication of incomprehensible data. But in data hiding, data is hidden in a cover file and it will be transmitted over the network. Hiding the existence of secret information is the main merit of data hiding techniques over cryptography [1,2,3].

The some types of data hiding approaches are Watermarking, Steganography and Reversible Data Hiding (RDH). Watermarking is a sequence of digital bits placed in a digital cover file that recognize the file's copyright information [4]. Steganography is dedicated for covert communication. It changes the image in such a way that only the sender and the intended receiver can detect the message while sending through it. Since it is invisible, the detection of secret data is not to be considered as simple. In steganography, the cover image does not hold any significance after extraction of secret data. Whereas in RDH the cover file also holds the secret data. The information of a patient that need to be protect such as patient's personal details, medical histo ry (includes past test results) and all current test reports.

The methods used to protect patient's information are classified in two broad ways.

* To maintain the patient's privacy, the patient's ID with the reports shouldn't be incorporated. So, an unauthorised person cannot recognize the identity of the patient.

* Even if the attacker gains access to the reports, it is difficult to interpret the report while encrypting the whole data. It relies on the strength of the encryption algorithm and proper management of the key between concerned parties such as the doctor, patient, etc..

Even though these methods work well, they have certain limitations. If the identity is missed or not specified, then it looks enveloping the particular report. Take for instance, the structure of a filename which identifies the patient. The document itself has no information about identifying the patient but the filename does. If the filenames of two files were swapped, then the reports themselves are also swapped. This can lead to wrong diagnosis of both patients. Encrypting the medical information which contains the identity would seem better.

Thus, both have their strengths and weakness. However, the first method can still be improved by not associating the identification "directly". Using a data hiding scheme, the identification of data can be embedded in the report. It makes sure that the report itself contains the information about a patient while it has not disclosing it directly. During diagnoses, patient data can be retrieved and check which patient it is for. Furthermore, the report with embedded data can be encrypted for stronger security. To cause danger attacker would need to decrypt the information and extract the hidden data.

Thus, it is established that both cryptography and steganography together can used to protect medical information in digital form. Cryptography used to encrypt the reports and steganography is used to embed the identification.

II. Basics Of Steganography:

Steganography or the concept of data hiding was first mentioned in a work by Johannes Trithemus (14621516) titled "Steganographia". The word "Steganography" is drieved from two Greek words "Steganos" and "graphia" meaning "Covered" and "Writing" Steganography has been used over the centuries. It is documented that in Demaratus sent a warning to the Spartans using steganography to initimate the allies that Xerxe's army approaching their country for war. According to the modern world, the idea of information hiding shown in FIG 1 or steganography was initially presented with the case of prisoner's mystery message by [15] and [13]

Process of data hiding:

The process of data hiding could be classified into three stages namely embedding stage, attacking stage, extracting stage as shown in FIG 2. In embedding stage, the embedding algorithm and the secret key is used to embed the secret data in the cover image. Then, the stego image is crested and transmitted over the network. In attacking stage, there is a possibility that either someone can attack the stego image or it gets corrupted by some noise i.e., Stego image is either altered or destroyed [16]. In extracting stage, the secret data is extracted from stego image by using same algorithm and secret key which is used in the embedding stage.

To design a perfect data hiding system, the following factors are to be considered:

* Imperceptibility: It is a technique, where the information is undetected by the Human Visual System (HVS).

* Security: It is the resistance of the technique to an attack even after realization of the existence of secret data.

* Capacity (Payload): Without affecting the visual quality the data can be concealed in the cover image.

* Robustness: To oppose unintentional actions like filtering, cropping, rotation, compression, etc.. in the stego image.

* Embedding complexity: It measures the complexity of the data embedding algorithm.

Different types of digital objects like text, image, audio and video are popular as cover files in data hiding [14]. Text data hiding lacks in security and embedding capacity. Any small change in the audio and video files in a moving stream of information is noticed. To embed secret data, image data hiding provides acceptable static redundant information. Therefore, the images are the most commonly used file format. Data hiding in digital images is a rapidly growing research area.

Image steganography is used to increase the medical image security, confidentiality and integrity. Medical image steganography is a special subcategory of image steganography where, the images have special requirements. Particularly, steganographed medical images should not differ perceptually from their original counterparts, because the clinical reading of the images must not be affected. Patient medical history is moving onto digital media with high resolution images like CT-SCAN and X-RAY results. However, this introduce new threats such as breach of privacy and tampering of results.

The privilege of 'Doctor-Patient confidentiality' has given to determine how and when their health information should be shared. If such information is disclosed to an employer, insurer, or the general public, it can result in discrimination or embarrassment. Thus, it is important to archive such information while maintaining confidentiality, integrity and availability. Additionally, when hard copy reports are given to doctors, they are enveloped with the patient ID (if not directly written on the report itself). In such cases, there is a chance of two reports being swapped. Since, with medical information both cryptography and steganography can be used to protect patient information as well as the medical results.

III. Methods Used In Medical Image Steganography:

We present some of the techniques used in medical image steganography and the comparison between them. Spatial domain technique is done by changing the bits of the binary representation of the medical image. Frequency domain or Transform domain technique is done by transforming the image to frequency. The classification of the medical image steganography is shown in FIG 3.

Spatial domain Steganography Techniques:

i) LSB based Steganography:

The LSB [6] is one of the first algorithms proposed for data hiding. It embeds the secret data in the LSBs of the cover image. The contribution of the LSB of the pixel to the image will be very less compared to the other bit. Here, the LSB of each byte of the cover image (Original medical image) will be replaced with the secret message. Only the LSB plane of the entire image will be affected, which does not causes much distortion to the cover image.

For example, 240 can be hidden in the first eight bytes of three pixels in a 24 bit image.
PIXELS: 00100111 11101001 11001000
        00100111 11001000 11101001
        11001000 00100111 11101001


Here, if we want to hide 240 in a medical image the first step is to convert 240 into binary number that is 011110000 then this 9 bit data is replaced by each LSB of the pixels of the image.
RESULT: 00100110 11101001 11001000
        00100111 11001001 11101000
        11001000 00100110 11101000


Because of sequential embedding the primitive LSB embedding techniques are insecure. To overcome this limitation the random embedding techniques were used for better security [17]. While maintaining better PSNR, it supports high embedding capacity. But any modification in the stego image leads to modification in patient data. Hence, robustness of these method is very low.

ii) Compression based RDH Steganography:

In this approach, spatial domain compression techniques are applied on bit planes of the cover image to generate space for secret data embedding [9]. A Generalised LSB (GLSB) [5] embedding is used to quantize the image and lossless compression technique is used to calculate the difference between quantized pixel value and cover pixel value. The compression provides some empty space to store the secret data. It is not robust against intentional attacks.

iii) Edge-Adaptive Steganography:

In this approach, to avoid the smooth regions the position is carefully chosen to embed the patient data. Only the sharper edge regions are selected to embed the secret data. Thus, the quality of the embedded image will be better than LSB method. It is again the vulnerability to attack [18] i.e., by changing the LSB of all pixels, an attacker can destroy the secret.

iv) PVD based Steganography:

The PVD divides the cover image into collection of non-overlapping two pixel blocks and finds the difference between each block. The range- table determines the amount of patient data that has to be embedded in each block. Data bits are embedded by altering the pixel block values such that the difference lies in the same range after modification [19]. There are some targeted attacks which exploits the difference in histogram.

v) BPCS based Steganography:

BPCS make use of human vision where the vessel image (cover image) is divided into "informative region" and "noise-like region" and the patient data is hidden in noise blocks of vessel image without degrading image quality [10,12]. The patient data is hidden in Most Significant Bit (MSB) planes along with the LSB planes. It is computationally complex, but the robustness, imperceptibility is achieved.

Transform domain Steganography Techniques:

vi) DCT based Steganography:

Discrete Cosine Transform (DCT) is used to converts spatial pixel intensities into Alternate Current (AC) and Direct Current (DC) coefficients. It is observed that most of the DCT based data hiding process use JPEG compression model. Initially, the cover image would be divided into non-overlapping blocks each of size 8*8 shown in FIG 4. Embedding algorithm alters the quantized coefficients according to the medical data. Blocking artifacts is main problem in DCT, it degrades the visual quality of the reconstructed medical image [11].

vii) DWT based Steganography:

In DWT based LSB embedding, the secret data bits are stored in the LSB positions of the quantized DWT sub-band coefficients is shown in FIG 5. It classifies the input signal into frequency ranges and supports the multi-resolution analysis. It embeds the compressed data along with the patient data in the high frequency bands [6,8. The DWT converts an image into floating point coefficients in the transform domain. But it would reduce the embedding capacity.

IV.Analysis And Recommendations:

The basic characteristics of a data hiding techniques are security, robustness, capacity and imperceptibility. But these parameters conflict each other. Table 1 provides the overall performances of data hiding techniques in both domains. Most of the researchers are taking extra care on imperceptibility and payload as compared to robustness. But robustness is essential to protect the secret data against various geometrical and image processing attacks.

Conclusion:

The study presented the significant of the medical image security during transmission. In order to increase the security of the stego image the cover image can be encrypted before embedding the medical data. Hence, more attention is needed to increase the robustness of the embedding algorithm. An implementation of hybrid method would be better than the existing data hiding schemes. So, the new method need to be developed with more data hiding capacity and against to the resistant of attack.

REFERENCES

[1.] Adams, M.D. and F. Kossentni, 2000. Reversible integer-to-integer wavelet transforms for image compression: performance evaluation and analysis. IEEE Trans. Image Proc., 9: 1010-1024.

[2.] Alattar, A.M., 2003. Reversible watermark using difference expansion of triplets. Proceedings of the International Conference on Image Processing, 1: 501-504.

[3.] Almohammad, A., R.M. Hierons and G. Ghinea, 2008. High capacity steganographic method based upon JPEG. Proceedings of the 3rd International Conference on Availability, Reliability and Security, pp: 544549.

[4.] Alsaade, F.W., 2016. Watermarking system for the security of medical image databases used in telemedicine. Res. J. Inform. Technol., 8: 88-97.

[5.] Celik, M.U., G. Sharma, A.M. Tekalp and E. Saber, 2005. Lossless generalized-LSB data embedding. IEEE Trans. Image Process., 14: 253-266.

[6.] Chan, C.K. and L.M. Cheng, 2004. Hiding data in images by simple LSB substitution. Pattern Recognit., 37: 469-474.

[7.] Chan, Y.K., W.T. Chen, S.S. Yu, Y.A. Ho, C.S. Tsai and Y.P. Chu, 2009. A HDWT-based reversible data hiding method. J. Syst. Software, 82: 411-421.

[8.] Chandramouli, R., M. Kharrazi and N. Memon, 2003. Image steganography and steganalysis: Concepts and practice. Proceedings of the 2nd International Workshop on Digital Watermarking, October 20-22, Seoul, South Korea, pp: 35-49.

[9.] Fridrich, J., M. Goljan and R. Du, 2002. Lossless data embedding-new paradigm in digital watermarking. EURASIP J. Applied Signal Process., pp: 185-196.

[10.] Hideki Noda, Jeremiah Spaulding, Mahdad N. Shirazi, and Eiji Kawaguchi, 2002. Application of Bit-Plane Decomposition Steganography to JPEG2000 Encoded Images. IEEE Signal Processing Letters, 9: 12.

[11.] Huang, F., X. Qu, H.J. Kim and J. Huang, 2016. Reversible data hiding in JPEG images. IEEE Trans. Circ. Syst. Video Technol., 26: 1610-1621.

[12.] M. Niimi, H. Noda and E. Kawaguchi, 1998. A steganography based on region segmentation by using complexity measure (in Japanese). Trans. IEICE, vol. J81-D-II, pp: 1132-1140.

[13.] Petitcolas, F.A.P., R.J. Anderson and M.G. Kuhn, 1999. Information hiding-a survey. Proc. IEEE, 87: 1062-1078.

[14.] Saha, B. and S. Sharma, 2012. Steganographic techniques of data hiding using digital images. Defence Sci. J., 62: 11-18.

[15.] Simmons, G.J., 1984. The Prisoners' Problem and the Subliminal Channel. In: Advances in Cryptography. Chaum, D. (Ed.). Springer, New York, USA., ISBN-13: 9781468447323, pp: 51-67.

[16.] Thanikaiselvan, V., S. Shastri and S. Ahmad, 2017. Information Hiding: Steganography. In: Intelligent Techniques in Signal Processing for Multimedia Security, Dey, N. and V. Santhi (Eds.). Springer, Germany, ISBN-13: 978-3319447896, pp: 65-91.

[17.] Wang, R.Z., C.F. Lin and J.C. Lin, 2001. Image hiding by optimal LSB substitution and genetic algorithm. Pattern Recognit., 34: 671-683.

[18.] Weiqi Luo, Fangjun Huang, and Jiwu Huang, 2010. Edge Adaptive Image Steganography Based on LSB Matching Revisited. IEEE Transactions on Information Forensics and Security, 5: 2.

[19.] Wu, D.C. and W.H. Tsai, 2003. A steganographic method for images by pixel-value differencing. Pattern Recognit. Lett., 24: 1613-1626.

(1) G. Santhi and (2) B. Adithya

(1) Assistant Professor, Department of IT, Pondicherry Engineering College, Puducherry, India.

(2) M.Tech (IT), Pondicherry Engineering College, Puducherry, India.

Received 14 September 2017; Accepted 15 October 2017; Available online 30 October 2017

Address For Correspondence:

G. Santhi, Assistant Professor, Department of IT, Pondicherry Engineering College, Puducherry, India.

E-mail: shanthikarthikeyan@pec.edu

Caption: Fig. 1: Basic Steganography

Caption: Fig. 2: Process of data hiding

Caption: Fig. 3: Classification of medical image steganography

Caption: Fig. 4: DCT based data hiding

Caption: Fig. 5: DWT based data hiding
Table 1: Performance evaluation of image data hiding techniques

Method                  Data Hiding   Resistance
                        Capacity      to Attacks

LSB based               High          Low
  Steganography
Compression based RDH   Low           Moderate
  Steganography
Edge-Adaptive           Average       Low
  Steganography
PVD based               High          Low
  Steganography
BPCS based              High          High
  Steganography
DCT based               Moderate      High
  Steganography
DWT based               Low           High
  Steganography

Method                  Domain      Complexity

LSB based               Spatial     Simple
  Steganography
Compression based RDH   Spatial     Complex
  Steganography
Edge-Adaptive           Spatial     Simple
  Steganography
PVD based               Spatial     Simple
  Steganography
BPCS based              Spatial     Complex
  Steganography
DCT based               Frequency   Complex
  Steganography
DWT based               Frequency   Complex
  Steganography
COPYRIGHT 2017 American-Eurasian Network for Scientific Information
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Santhi, G.; Adithya, B.
Publication:Advances in Natural and Applied Sciences
Article Type:Report
Date:Oct 1, 2017
Words:2692
Previous Article:Preserving security and privacy in mobile cloud computing health care service w.r.t noncontact ECG monitoring.
Next Article:A Survey on Big Data Privacy and Security Issues in Healthcare Information System.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |