A Fuzzy Logic based Bit Plane ComplexitySegmentation steganography to secure Electronic Health RecordsG. Santhi, B. Adithya InformationTechnology, Pondicherry [email protected], [email protected] Abstract— Electronic Healthcare or E-Healthcare (EH) is apaperless management promises to speed up typical bureaucracy of healthcare.
Atypical E-Healthcare system consists of many component and subsystem, such asappointment, routine clinical notes, picture archiving, lab and radiologyorders, etc., are vulnerable to security threats. Cryptology and steganographyare generally used to ensure medical data security. For this reason, a FuzzyLogic based Bit Plane Complexity Segmentation (FL-BPCS) steganography is combinedwith AES cryptography and Huffman lossless compression is proposed to securepatient data’s. In this, Electroencephalogram time series, doctor’s comment andpatient information are selected as hidden data and Magnetic Resonance imageare used as cover image. Keywords—E-Healthcare,steganography, FL-BPCS, Electroencephalogram, cryptology I.
INTRODUCTIONThe privacy andsecurity of EH information falls into two categories. First, inappropriarereleases from authorized users who intentionally or unintentionally access ordisseminare information violation of EH system policies or EH computer systemcan be break by outsiders. Second, open disclosure of patient healthinformation to parties against to the interests of specific individual patientor invadere a patient’sprivacy.These falls arises from the flow of data across the EH system amongst andbetween providers, payers and secondary users, with or without the patient’sknowledge are conceptually quite differ or require different counteracts andinterventions.Technicalobstaculum of the intrudere includes the use of firewalls to isolate internalnetworks with strong encryption based authentication and authorization. Thereis no known obstaculum for external networks like Denial-of-Service. Nevertheless,technical countermeasures cannot be cures all security threats.
Obstaculum suchas encryption, steganography and authentication are the only effective ways tocounter EH security threats against internet interface. Steganography isused for covert communication 2. The embedding algorithm will convert thecover medium into stego medium by embedding secret data into it. The inverseprocess of embedding is done to extract the secret data. Imperceptibility,security, capacity, robustness, embedding complexity are the steganographyfactors that has to be considered. Image steganography is developed accordingto its use in medical fields to communicate between patient as well ascommunication between doctor’s and laboratory people’s to hide secret messages.
Steganography is to avoid drawing attention to the transmission of hiddenmedical information. If suspicion is raised, steganography and cryptography isplanned to achieve the security of secret medical data’s. Both are complementaryto each other and provide better security, confidentiality and authenticity.Eiji Kawaguchiet al. proposed a Bit Plane Complexity Segmentation (BPCS) to increase theembedding capacity and also to overcome the short comes of traditionalsteganography techniques such as Least Significant Bit (LSB), Transformembedding, Perceptual masking techniques. But the issue is embedded secret datacan be retrieved by using Difference Image Histogram (DIH) 8.Karakis et al.
proposed a Fuzzy Logic based Least Significant Bit (FL-LSB) to reduce theinsecurity of LSB planes. By using fuzzy the LSB planes were chosen to embedthe secret data’s. But the issue, is low embedding capacity and the stego imageis invalid 9. This technique can be offered by using the following sections.In section 2, the brief description of BPCS, cryptology as well as lossless compressiontechnique is presented, in section 3, the work methodology is presented, insection 4, results are presented, in section 5, the discussion and analysis aredone, and finally the work is concluded.
II. BRIEFDESCRIPTION OF BPCS, CRYPTOLOGY AND LOSSLESS COMPRESSION TECHNIQUE A.Bit Plane Complexity Segmentation for embed and extract Thissteganography method makes use of the human vision. The cover image is dividedinto informative region and noise-like region and the secret data is hidden innoise blocks of vessel image without degrading image quality. The data ishidden in both Most Significant Bit (MSB) as well as LSB planes 8.
B.Cryptologywith security concern Cryptology usesthe Advanced Encryption Standard (AES) to encrypt the secret data to formsecond security layer. It comprises of a series of linked operations, some ofwhich involve replacing the inputs by specific outputs (Substitution) and othersinvolve shuffling bits around (Permutations). It treats the 128 bits of aplaintext block as 16 bytes as a matrix. It has built-in flexibility of keylength, which allows a degree of future-proofing against progress in theability to perform exhaustive-key searches 4. C.Losslesscompression technique to reduce the size Losslesscompression technique uses the Huffman to reduce the secret data size.
Thismethod takes a symbol (bytes) and encodes them with variable length codesaccording to the statistical probabilities. A frequently used symbol will beencoded with couple of bits, while symbol that are rarely used will be encodedwith more bits 1. III. PROPOSEDWORKThe proposed system aims to use EEGsignals and MR images that are obtained from same patients and to embed moredata with EEG into MR images of same patient.
For this reason, the MR imagesand EEG of epilepsy patients are gathered from the Department of Neurology atBonn University. 12 females and 11 males were included (age: 18-65 years; meanage: 35 ± 7.7 years). The embedded message was combined with the patient’sinformation, doctor’s comments, and EEG time series from the EEG file header.
The patient’s information (patient name, patient ID, patient birth date,patient gender, patient age, patient weight, patient address, studydescription, series date, series time, and series description) were separatelyselected from the meta-header of each of the DICOM files. Least Significant Bit (LSB) and MostSignificant Bit (MSB) embedding is a simple and fast strategy in steganography.It has high imperceptibility and embedding capability. Hence, this studyproposes new methods to modifyLSB and MSB embedding using medical data.
The analyses consist of two stages:embedding and extracting, respectively. Initially, the patient’s information isobtained from DICOM series of epilepsy patient. The EEG data is segmentedaccording to the size of these DICOM images. An image is sampled by pixels. In thegray-scale image, pixels have gray level intensities.
In color images, pixelsare also represented by three component intensities, being red (R), green (G),and blue(B). A similarity measure is the similarity degree between two groups orbetween two objects. In image processing, the similarity measure of two pixelsis used with distance information in Euclidean color space. Demirci 6proposed a similarity-based method for edge detection. Furthermore, Pixel-ValueDifferencing (PVD) or Adjacent Pixel Difference (APD) methods determineembedding pixels in histogram-based steganography. Data Flow of overallproposed system process is shown in Fig. 1.
These methods have high embeddingcapacity and PSNR values. The main idea of this method is togenerate a new image whose pixels have double values at the interval 0 1.This similar image of cover image is used to determine pixels for the embeddingmessage. In this method, if the values of similar pixels are higher than thedetermined threshold by trial and error, they are selected to hide the message9.
The neighboring pixels of the image (P1,P2,…P9) using the 3×3 windowhave three color component (R, G, B).The gray leveldifferences of color components are calculated the neighboring pixels of thestego image. The color distance of pixels are calculated by the Euclidean norm9. Thesimilarity values of pixels are founded.
Similarly, the coordinates ofthe pixels are determined between the similarity values of pixels and thresholdvalues. The hidden message is extracted using the coordinates of the stegoimage’s pixels. Fig. 1 Data flow of theoverall proposed work In message pre-processing stage,lossless compression techniques, which are LZW (Lempel–Ziv–Welch) and HuffmanCompression, are used to increase message capacity. Furthermore, LZW andHuffman Compression methods also ensure the complexity of the message.
To increase security, the compression message is encrypted by the Rijndaelsymmetric encryption algorithm using a 128-bit key. Secondly, the proposedmethods, which are Fuzzy-Logic-based Bit Plane Complexity Segmentation (FL-BPCS),select MSBs and LSBs of image pixels with using the differences in gray levelsof the pixels. Finally, the selected MSBs and LSBs of the pixels are alteredwith the message bits in stego images. These processes are simultaneously runwith all DICOM series to decrease computational time. The extracting messagestage requires stego-DICOM images is shown in Fig.
2a, 2b and a stego-key,which is the authentication key for decryption. Firstly, the proposed methodsgive the pixels coordinates, which have an embedded message. These pixels areused to gather the message.
Secondly, the obtained message is decrypted anddecompressed. Finally, the patient’s information, segmented EEG, and thedoctor’s comments are displayed in a GUI (Graphical User Interface) screen. Allhidden EEG data can be also gathered from the DICOM series.
The comparisonresults of the proposed algorithm are evaluated by Histogram is shown in Fig.3a, 3b, PSNR (peak signal-to-noise ratio), MSE (mean square of error), SSIM(structural similarity measure), between the cover, and the stego-DICOM series. IV. RESULTSDISCUSSION AND ANALYSISThe proposed system parametersare used to evaluate the performance of the data hiding techniques.Peak Signal to Noise Ratio (PSNR): The PSNR is generally usedto measure the quality of stego image in decibels (dB). Eq.1, gives the expression forPSNR in which ICmax isthe maximum pixel value of the cover image and MSE is the mean square error:Where,In Eq.
2, x and y are the image coordinates, M and N arethe dimensions of the image, ISxy isthe generated stego-image and ICxy isthe cover image 7,10.Structural similarity (SSIM) index: The SSIM is a method for findingthe similarity between cover image and the stego image. It is aperception-based model that considers image degradation as perceived change instructural information. The SSIM measure between two images and is represented in Eq. 3, where, is the average , is the average of is the variance of is the variance of is the covariance between and and k1, k2 aretwo the variables used to stabilize the division with weak denominator 10. (a) (b)Fig.
2: (a)Cover image(b)Stego image (a) (b)Fig. 3: Histogram of (a)Coverimage (b)Stego imageThegraph is created based on the embeddingcapacity is shown in Fig. 4. For embedding payload, we use an embeddingrate, to represent the percentage of the embedded secret bits in the wholepixels of the cover image 7. In Fuzzy Logic based Least Significant Bit theembedding capacity is mentioned as low, because to embed the secret data itoccupies only the LSB positions.
In Bit Plane Complexity Segmentation theembedding capacity is mentioned as low, when compared to the Fuzzy Logic basedLeast Significant Bit but normally the embedding capacity is high when comparedto the primitive LSB because to embed the secret data it occupies both MSB aswell as LSB. In Fuzzy Logic based Bit Plane Complexity Segmentation theembedding capacity is high because to embed the secret data the red, green,blue channel were used with fuzzy.Fig.
4:Embedding capacity Thegraph is created based on the performance evaluation parameters PSNR, MSE, SSIM is shown in Fig. 5. Peak signal-to-noise ratio,often abbreviated PSNR, isan engineering term for the ratio between the maximum possible power of a signal and the power ofcorrupting noise thataffects the fidelity of its representation. PSNRis most easily defined via the meansquared error (MSE). Two of the error metrics used to compare the various image compression techniques arethe Mean Square Error (MSE) and the Peak Signal to NoiseRatio (PSNR).
The MSE is the cumulative squared error between thecompressed and the original image,whereas PSNR is ameasure of the peak error 3. Structural Similarity Index (SSIM) is a perceptual metricthat quantifies image qualitydegradation caused by processing suchas data compression or by losses in data transmission. It is a full referencemetric that requires two images from the same image capture— a reference image and a processed image5. InFuzzy Logic based Least Significant Bit the PSNR is achieved high, because toembed the secret data it occupies only the LSB positions and MSE is achievedlow when compared to Bit Plane Complexity Segmentation and SSIM is achievedhigh. In Bit Plane Complexity Segmentation the PSNR is achieved as low becauseto embed the secret data it occupies both MSB as well as LSB and MSE isachieved high and SSIM is low because MSB is more sensitive. While changingthat pixel values with another pixel values must be matched. In Fuzzy Logicbased Bit Plane Complexity Segmentation the PSNR is low because to embed thesecret data MSB as well as LSB is used. MSE is low and SSIM is high because ituses red, green, blue channels with fuzzy rules.
Fig. 5:Performance ParametersThegraph is created based on the steganalysisis shown in Fig. 6. Visualattacks involve observing the unusual patterns and noisy blurred regions insome places of the stego image.
Astatistical method called RS steganalysis for detection of LSB embedding usesdual statistics derived from spatial correlation of an image. Histogram based steganalysistechniques detect the existence of secret data from smoothness of the stegoimage histogram. Similarly, a targeted active steganalysis technique isimplemented for HS embedding using the change in the characteristics ofhistogram during data embedding 10. In Fuzzy Logic based Least Significant Bit thevisual attacks, RS statistical attack, Sample Pair Analysis is achieved high, because to embed the secret data itoccupies only the LSB positions but Difference Image Histogram is achieved low.In Bit Plane Complexity Segmentation the visual attacks, RS statistical attack,Sample Pair Analysis is achieved low, because to embed the secret data itoccupies both MSB as well as LSB but Difference Image Histogram is achievedhigh.
In Fuzzy Logic based Bit Plane Complexity Segmentation the visualattacks, RS statistical attack, Sample Pair Analysis is achieved low, becauseto embed the secret data the red, green, blue channel were used with fuzzyrules but Difference Image Histogram is achieved high. Fig. 6:Detection of Embedded Data’sV. CONCLUSIONIn medical information system, medicaldata is easily captured when being storing, receiving or transmission throughcomputer network and Internet.
Cryptology and steganography are generally usedto ensure medical data security. For this reason, this study proposes newalgorithm, Fuzzy Logic-based Bit Plane Complexity Segmentation (FL-BPCS) tosecure medical data. EEG signals and MR images of epilepsy patients are used tocombine multiple medical signals into one file format. The embedding messagesare composed of EEG signals, doctor’s comment, and patient information in fileheader of DICOM images.
The messages are secured by using Huffman losslesscompression methods and Rijndael symmetric algorithm with 128 bit-key toprevent the attacks. The capacity of proposed algorithm ishigher than the result of similar studies in literature. According to theobtained result, the proposed method ensures the confidentiality of thepatient’s information. The FL-BPCS method hides EEG signals, patient’sinformation and doctor’s comment in the pixels of MR images. It also reducesdata repository and transmission capacity of the patients’ multiple medicaldata. In future, the embedding and extracting of medical data from an cloudstorage with multiple records can be implemented for flexibility to access thepatient data for diagnosis. To access the patient data the grant privilege isgiven to make read and write permission for authorized patient’s as well as doctor’s,nurse with laboratory people’s. REFERENCES1 Huffman, D.
, “A Method for the Construction of Minimum-Redundancy Codes”, Proceedings of the IRE. 40 (9): 1098–1101, 1952. doi:10.
1109/JRPROC.1952.2738982 Simmons, G.
J., “The Prisoners’ Problem and the Subliminal Channel”,In: Advances in Cryptography, Chaum, D. (Ed.). Springer, New York, USA.,ISBN-13: 9781468447323, pp: 51-67, 1984.
3 Welstead, Stephen T,” Fractaland wavelet image compression techniques”, SPIEPublication, pp. 155–156, 1999. ISBN 978-0-8194-3503-34 Daemen, Joan; Rijmen, Vincent, “AES Proposal:Rijndael”, National Institute of Standards and Technology, p. 1. March 9, 2003.
5 Wang, Zhou; Bovik, A.C.; Sheikh, H.R.; Simoncelli,E.
P., “Image quality assessment: fromerror visibility to structural similarity”, IEEETransactions on Image Processing, 13 (4): 600–612,2004-04-01. doi:10.
1109/TIP.2003.8198616 RecepDemirci, “Similarity relation matrix-based color edge detection”,AEU-International Journal of Electronics and Communications, vol 61, issue 7,pages: 469-477, 2007.7 Yanping Zhang, Juan Jiang, Yongliang Zha, HengZhang, Shu Zhao, “Research on Embedding Capacity and Efficiency of InformationHiding Based on Digital Images”, SciRes.., International Journal ofIntelligence Science, 77-85, 2013.
8 SouvikBhattacharyya, Aparajita Khan,Aunkita Nandi,Aveek Dasmalakar,Somdip Roy andGautam Sanyal, “Pixel Mapping Method (PMM) Based Bit Plane ComplexitySegmentation (BPCS) Steganography”, Informationand Communication Technologies,pages: 36-41, 2014.9 R.Karak??, ?. Güler, ?.
Çapraz and E. Bilir, “A Novel Fuzzy Logic-Based ImageSteganography Method To Ensure Medical Data Security”, Computers in Biology and Medicine,vol 67, pages: 172-183, 2015.10 AhmadShaik, V.
Thanikaiselvan and Rengarajan Amitharajan,. “Data Security Through DataHiding in Images: A Review”, Journalof Artificial Intelligence, 10: 1-21, 2017.