You are on page 1of 4

Bonfring International Journal of Software Engineering and Soft Computing, Vol. 9, No.

2, April 2019 43

Block Chain Technology for Privacy Protection for


Cloudlet-based Medical Data Sharing
M. Divya and Dr.G. Singaravel

Abstract--- With the popularity of wearable devices, along be the driving technology behind the next generation of
with the development of clouds and cloudlet technology, there Internet, also referred to as the Decentralized Web, or the
has been increasing need to provide better medical care. The Web3. Blockchain is a novel solution to the age-old human
processing block chain of medical data mainly includes data problem of trust. It provides an architecture for so-called
collection, data storage and data sharing, etc. Traditional trustless trust. It allows us to trust the outputs of the system
healthcare system often requires the delivery of medical data without trusting any actor within it.
to the cloud, which involves users’ sensitive information and A Blockchain protocol operates on top of the Internet, on a
causes communication energy consumption. Practically, P2P Network of computers that all run the protocol and hold
medical data sharing is a critical and challenging issue. Thus
an identical copy of the ledger of transactions, enabling P2P
in this paper, we build up a novel healthcare system by value transactions without a middleman though machine
utilizing the flexibility of cloudlet. The functions of cloudlet consensus. Blockchain itself a file – a shared and public ledger
include privacy protection, data sharing and intrusion of transactions that records all transactions from the genesis
detection. In the stage of data collection, we first utilize block (first block) until today. Blockchain is a shared, trusted,
Number Theory Research Unit (NTRU) method to encrypt
public ledger of transactions, that everyone can inspect but
user’s body data collected by wearable devices. Those data which no single user controls. It is a distributed
will be transmitted to nearby cloudlet in an energy efficient database that maintains a continuously growing list of
fashion. Secondly, we present a new trust model to help users
transaction data records, cryptographically secured from
to select trustable partners who want to share stored data in tampering and revision.
the cloudlet. The trust model also helps similar patients to
communicate with each other about their diseases. Thirdly, we The ledger is built using a linked list, or chain of blocks,
divide users’ medical data stored in remote cloud of hospital where each block contains a certain number of transactions
into three parts, and give them proper protection. Finally, in that were validated by the network in a given timespan. The
order to protect the healthcare system from malicious attacks, crypto-economic rules sets of the blockchain protocol
we develop a novel collaborative intrusion detection system (consensus layer) regulate the behavioural rules sets and
(IDS) method based on cloudlet mesh, which can effectively incentive mechanism of all stakeholders in the network.
prevent the remote healthcare big data cloud from attacks. • Blockchain technology is like the internet in that it has
Our experiments demonstrate the effectiveness of the proposed a built-in robustness. By storing blocks of information
scheme. Combines classical partitioning algorithms with that are identical across its network, the blockchain
probabilistic models so as to form an efficient clustering cannot:
approach. • Be controlled by any single entity.
Keywords--- IDS, Cloudlets Mining, Number Theory, • Has no single point of failure.
Encrypt, Block Chain. Bitcoin was invented in 2008. Since that time, the Bitcoin
blockchain has operated without significant disruption. (To
date, any of problems associated with Bitcoin have been due
I. INTRODUCTION
to hacking or mismanagement. In other words, these problems

T HE client’s physiological data are first collected by


wearable devices such as smart clothing. Then, those data
are delivered to cloudlet. The following two important
come from bad intention and human error, not flaws in the
underlying concepts.)

problems for healthcare data protection is considered. The first The internet itself has proven to be durable for almost 30
problem is healthcare data privacy protection and sharing data. years. It’s a track record that bodes well for blockchain
The second problem is to develop effective countermeasures technology as it continues to be developed.
to prevent the healthcare database from being intruded from B. Content Sharing and Privacy Protection
outside. Introduce the encryption process for users’ privacy data,
A. Block Chain which prevents the leakage or malicious use of users’ private
Blockchain, the technology behind Bitcoin, seems to data during transmissions. Next, we present the identity
management of users who want to access to the hospital’s
healthcare data. Thus, we can assign different users with
M. Divya, PG Scholar, Department of Information Technology, K.S.R.
College of Engineering, Tiruchengode, India. different levels of permissions for data access, while avoiding
Dr.G. Singaravel, Professor & Head, Department of Information data access beyond their permission levels. Finally, we give an
Technology, K.S.R. College of Engineering, Tiruchengode, India. application of using users’ private data, which is beneficial to
DOI:10.9756/BIJSESC.9021

ISSN 2277-5099 | © 2019 Bonfring


Bonfring International Journal of Software Engineering and Soft Computing, Vol. 9, No. 2, April 2019 44

both users and doctors. Based on the healthcare big data stored complete the intrusion detection task. We use {S1, S2, . . . ,
in the remote cloud, a disease prediction models built based on Sm} to represent the set of IDS’s in the IDS (CIDS) .Suppose
decision tree. The predictions will be reported to the users and that each IDS is able to detect intrusion independently. For the
doctors on demand. sake of simplicity, we use I to indicate that there is intrusion
behaviour.
C. Encryption at the User End
When using wearable devices to collect users’ data, the II. RELATED WORKS
procedure inevitably involves the user’s sensitive information.
Therefore, how to effectively collect and transmit users’ data M. S. Hossain et al., [1] the potential of cloud-supported
under efficient privacy protection is a critical problem. In a cyber-physical systems (CCPSs) has drawn a great deal of
data collection method, called PHDA, is proposed based on interest from academia and industry. CCPSs facilitate the
data priority which can give proper cost and delay to different seamless integration of devices in the physical world (e.g.,
priorities data. In, Li et al. discuss the process of data sensors, cameras, microphones, speakers, and GPS devices)
collection and utilizes sum aggregation to obtain data to make with cyberspace. This enables a range of emerging
sure the security of users’ privacy in the presence of unreliable applications or systems such as patient or health monitoring,
sensors. In , Lu et al., study 3V data privacy protection issue which require patient locations to be tracked. These systems
based on big data of healthcare. Based on the model presented integrate a large number of physical devices such as sensors
in, this paper utilizes the advantages of NTRU encryption with localization technologies (e.g., GPS and wireless local
scheme. NTRU can protect the user’s physiological data, such area networks) to generate, sense, analyse, and share huge
as heart rate, blood pressure and Electrocardiography (ECG), quantities of medical and user-location data for complex
etc. Before transmitted to a smartphone, NTRU encryption processing. However, there are a number of challenges
scheme executed. The encrypted data will then be stored in the regarding these systems in terms of the positioning of patients,
cloudlet through a cellular network or WiFi. We hereby ubiquitous access, large-scale computation, and
describe the processes of encryption and deciphering in the communication. Hence, there is a need for an infrastructure or
following. system that can provide scalability and ubiquity in terms of
huge real-time data processing and communications in the
cyber or cloud space. To this end, this paper proposes a cloud-
supported cyber-physical localization system for patient
monitoring using smartphones to acquire voice and
electroencephalogram signals in a scalable, real-time, and
efficient manner. The proposed approach uses Gaussian
mixture modelling for localization and is shown to outperform
other similar methods in terms of error estimation.
K. T. Pickard and M. Swan [2] Sharing personal health
information is essential to create next generation healthcare
services. To realize preventive and personalized medicine,
large numbers of consumers must pool health information to
In order to protect medical data, we also develop an create datasets that can be analysed for wellness and disease
intrusion detection system in this paper. Once a malicious trends. Incorporating this information will not only empower
attack is detected, the system will fire an alarm. This section consumers, but also enable health systems to improve patient
presents a novel scheme to build a collaborative IDS system to care. To date, consumers have been reluctant to share personal
deter intruders. In the following, we first consider what health information for a variety of reasons, but attitudes are
happens if the system is suffering from different attacks, while shifting. Results from an online survey demonstrate a strong
detection rates for individual IDS vary with the cloudlet willingness to share health information for research purposes.
servers. We will plot the detection rate and false alarm rate as Building on these results, the authors present a framework to
the receiver operating characteristic (ROC) curves. Next, we increase health information sharing based on trust, motivation,
evaluate the collaborative detection rate and estimate the community, and informed consent.
expected cost of implementation in the cloudlet mesh. We
apply a decision tree to choose the optimal number of IDS’s to J. Yang, J. Li, J. Mulder, Y. Wang [3] The appropriate
be deployed on the mesh. The goal is to achieve prescribed collection and consumption of electronic health information
detection accuracy against the false alarm rate under the about an individual patient or population is the bedrock of
premise of minimizing the system cost. modern healthcare, where electronic medical records (EMR)
serve as the main carrier. This paper first introduces the main
Collaborative IDS is designed among m IDS, S1, S2, Sm, goal of this special issue and gives a brief guideline. Then, the
in order to get higher detection rate and lower false alarm rate. present situation of the adoption of EMRs is reviewed. After
The IDS assumed to be detect Independently. There exists a K that, the emerging information technologies are presented
different type of intrusion. So according to deduce in the which have a great impact on the healthcare provision. These
following, we can get the detection rate and false alarm rate of include health sensing for medical data collection, medical
collaborative IDS. In order to evaluate it , we give the ROC data analysis and utilization for accurate detection and
curve. Before transmitting data to the remote cloud, we prediction. Next, cloud computing is discussed, as it may
establish the collaborative IDS based on the cloudlet mesh to

ISSN 2277-5099 | © 2019 Bonfring


Bonfring International Journal of Software Engineering and Soft Computing, Vol. 9, No. 2, April 2019 45

provide scalable and cost-effective delivery of healthcare individual distance metrics from each physician into a globally
services. Accordingly, the current state of academic research consistent unified metric? We describe a suite of supervised
is documented on emerging information technologies for new metric learning approaches that answer the above questions. In
paradigms of healthcare service. At last, conclusions are made. particular, we present Locally Supervised Metric Learning
Y. Wu, M. Su, W. Zheng, K. Hwang [4] In a community (LSML) to learn a generalized Mahalanobis distance that is
cloud, multiple user groups dynamically share a massive tailored toward physician feedback. Then we describe the
number of data blocks. The authors present a new associative interactive metric learning (iMet) method that can
data sharing method that uses virtual disks in the MeePo incrementally update an existing metric based on physician
cloud, a research storage cloud built at Tsinghua University. feedback in an online fashion. To combine multiple similarity
Innovations in the MeePo cloud design include big data measures from multiple physicians, we present Composite
metering, associative data sharing, data block prefetching, Distance Integration (Comdi) method. In this approach we
privileged access control (PAC), and privacy preservation. first construct discriminative neighbourhoods from each
These features are improved or extended from competing individual metrics, then combine them into a single optimal
features implemented in DropBox, CloudViews, and distance metric. Finally, we present a clinical decision support
MySpace. The reported results support the effectiveness of the prototype system powered by the proposed patient similarity
MeePo cloud. methods, and evaluate the proposed methods using real EHR
data against several baselines.
Kirti A. Dongre ; Roshan Singh Thakur [5] Cloud
computing is one of the upcoming technologies that will A. Sajid and H. Abbas, [8] The widespread deployment
upgrade generation of Internet. The data stored in the smart and utility of Wireless Body Area Networks (WBAN's) in
phones is increased as more applications are deployed and healthcare systems required new technologies like Internet of
executed. If the phone is damaged or lost then the information Things (IoT) and cloud computing, that are able to deal with
stored in it gets lost. If the cloud storage can be integrated for the storage and processing limitations of WBAN's. This
regular data backup of a mobile user so that the risk of data amalgamation of WBAN-based healthcare systems to cloud-
lost can be minimized. The user can stored data in the server based healthcare systems gave rise to serious privacy concerns
and retrieve them at anytime and from anywhere. The data to the sensitive healthcare data. Hence, there is a need for the
might be uncovered by attack during the retrieval or proactive identification and effective mitigation mechanisms
transmission of data using wireless cloud storage without for these patient's data privacy concerns that pose continuous
proper authentication and protection. So to avoid this in this threats to the integrity and stability of the healthcare
paper we design a mechanism that provides a security environment. For this purpose, a systematic literature review
requirement for data storage of mobile phones. has been conducted that presents a clear picture of the privacy
concerns of patient's data in cloud-assisted healthcare systems
X.Xue and W.B.Croft [6] propose a novel framework and analysed the mechanisms that are recently proposed by the
where the original query is transformed into a distribution of research community. The methodology used for conducting
reformulated queries. A reformulated query is generated by the review was based on Kitchenham guidelines. Results from
applying different operations including adding or replacing the review show that most of the patient's data privacy
query words, detecting phrase structures, and so on. Since the techniques do not fully address the privacy concerns and
reformulated query that involves a particular choice of words therefore require more efforts. The summary presented in this
and phrases is explicitly modelled, this framework captures paper would help in setting research directions for the
dependencies between those query components. On the other techniques and mechanisms that are needed to address the
hand, this framework naturally combines query segmentation, patient's data privacy concerns in a balanced and light-weight
query substitution and other possible reformulation operations, manner by considering all the aspects and limitations of the
where all these operations are considered as methods for cloud-assisted healthcare systems.
generating reformulated queries. In other words, a
reformulated query is the output of applying single or multiple Sachchidanand Singh ; Nirmala Singh[8] Blockchain is
reformulation operations. a decentralized ledger used to securely exchange digital
currency, perform deals and transactions. Each member of the
Jimeng Sun, Fei Wang, Jianying Hu, Shahram network has access to the latest copy of encrypted ledger so
Edabollahi[7] Patient similarity assessment is an important that they can validate a new transaction. Blockchain ledger is a
task in the context of patient cohort identification for collection of all Bitcoin transactions executed in the past.
comparative effectiveness studies and clinical decision support Basically, it's a distributed database which maintains a
applications. The goal is to derive clinically meaningful continuously growing tamper proof data structure blocks
distance metric to measure the similarity between patients which holds batches of individual transactions. The completed
represented by their key clinical indicators. How to blocks are added in a linear and chronological order. Each
incorporate physician feedback with regard to the retrieval block contains a timestamp and information link which points
results? How to interactively update the underlying similarity to a previous block. Bitcoin is peer-to-peer permission-less
measure based on the feedback? Moreover, often different network which allows every user to connect to the network
physicians have different understandings of patient similarity and send new transaction to verify and create new blocks.
based on their patient cohorts. The distance metric learned for Satoshi Nakamoto described design of Bitcoin digital currency
each individual physician often leads to a limited view of the in his research paper posted to cryptography listserv in 2008.
true underlying distance metric. How to integrate the Nakamoto's suggestion has solved long pending problem of

ISSN 2277-5099 | © 2019 Bonfring


Bonfring International Journal of Software Engineering and Soft Computing, Vol. 9, No. 2, April 2019 46

cryptographers and laid the foundation stone for digital IV. CONCLUSION
currency. This paper explains the concept, characteristics, In this paper, we investigated the problem of privacy
need of Blockchain and how Bitcoin works. It attempts to protection and sharing large medical data in cloudlets and the
highlights role of Blockchain in shaping the future of banking, remote cloud. We developed a system which does not allow
financial institutions and adoption of Internet of Things(IoT). users to transmit data to the remote cloud in consideration of
secure collection of data, as well as low communication cost.
III. METHODOLOGY However, it does allow users to transmit data to a cloudlet,
3.1 Client Data Encryption which triggers the data sharing problem in the cloudlet.
Firstly, we can utilize wearable devices to collect users’ data,
We utilize the model presented in, and take the advantage
and in order to protect users privacy, we use NTRU
of NTRU to protect the client’s physiological data from being
mechanism to make sure the transmission of users’ data to
leaked or abused. This scheme is to protect the user’s privacy
cloudlet in security. Secondly, for the purpose of sharing data
when transmitting the data from the smartphone to the
in the cloudlet, we use trust model to measure users trust level
cloudlet.
to judge level to share data or not. Thirdly, for privacy-
3.2 Cloudlet based Data Sharing preserving of remote cloud data, we partition the data stored in
Typically, users geographically close to each other connect the remote cloud and encrypt the data in different ways, so as
to the same cloudlet. It’s likely for them to share common to not just ensure data protection but also accelerate the
aspects, for example, patients suffer from similar kind of efficacy of transmission. Finally, we propose collaborative
disease exchange information of treatment and share related IDS based on cloudlet mesh to protect the whole system with
data. For this purpose, we use users’ similarity and reputation block chain technology. The proposed schemes are validated
as input data. After we obtain users’ trust levels, a certain with simulations and experiments.
threshold is set for the comparison. Once reaching or
exceeding the threshold, it is considered that the trust between REFERENCES
the users is enough for data sharing. Otherwise, the data will [1] M.S. Hossain, “Cloud-supported cyber–physical localization framework
not share with low trust level. for patients monitoring”, IEEE Systems Journal, Vol. 11, No. 1, Pp. 118-
127, 2015.
3.3 Remote Cloud Data Privacy Protection [2] K.T. Pickard and M. Swan, “Big desire to share big health data: A shift
in consumer attitudes toward personal health information”, In AAAI
Compared to user’s daily data in cloudlet, the data stored Spring Symposium Series, 2014.
in remote contain larger scale medical data, e.g., EMR, which [3] M.S. Hossain, G. Muhammad, M.F. Alhamid, B. Song and K. AlMutib,
will be stored for a long term. We use the methods presented “Audio-visual emotion recognition using big data towards 5g”, Mobile
Networks and Applications, Pp. 1–11, 2016.
to divide EMR into explicit identifier (EID), quasi-identifier [4] J.J. Yang, J. Li, J. Mulder, Y. Wang, S. Chen, H. Wu, Q. Wang and H.
(QID) and medical information (MI). After classifying, proper Pan, “Emerging information technologies for enhanced healthcare”,
protection is given for the data containing users’ sensitive Computers in Industry, Vol. 69, Pp. 3–11, 2015.
information. [5] D. Nunez, I. Agudo and J. Lopez, “Ntrureencrypt: An efficient proxy ˜
re-encryption scheme based on ntru”, In Proceedings of the 10th ACM
3.4 Collaborative Ids based on Cloudlet Mesh Symposium on Information, Computer and Communications Security,
There is a vast volume of medical data stored in the remote Pp. 179–189, 2015.
[6] Y. Wu, M. Su, W. Zheng, K. Hwang and A. Y. Zomaya, “Associative
cloud, it is critical to apply security mechanism to protect the big data sharing in community clouds: The meepo approach”, IEEE
database from malicious intrusions. In this paper, we develop Cloud Computing, Vol. 2, No. 6, Pp. 64–73, 2015.
specific countermeasures to establish a defence system for the [7] W. Xiang, G. Wang, M. Pickering and Y. Zhang, “Big video data for
large medical database in the remote cloud storage. light-field-based 3d telemedicine”, IEEE Network, Vol. 30, No. 3, Pp.
30– 38, 2016.
Specifically, collaborative IDS based on the cloudlet mesh [8] K. He, J. Chen, R. Du, Q. Wu, G. Xue and X. Zhang, “Deypos:
structure is used to screen any visit to the database as a Deduplicatable dynamic proof of storage for multi-user environments”,
protection border. If the detection shows a malicious intrusion IEEE Transactions on Computers, Vol. 65, No. 12, Pp. 3631-3645,
in advance, the collaborative IDS will fire an alarm and block 2016.
the visit, and vice-versa. The collaborative IDS, as a guard of
the cloud database, can protect a vast number of medical data
and make sure of the security of the database.

ISSN 2277-5099 | © 2019 Bonfring