Secure Electronic Health Record System Based on Online/Ofﬂine

KP-ABE in the Cloud

Kun Liu

School of Mathematics and Information Science, Guangzhou University, 230 Guangzhou University City Outer Ring Road,

Guangzhou, P.R. China

Keywords:

Electronic Health Record, Online/Ofﬂine Key-policy Attribute-based Encryption, Cloud Computing.

Abstract:

Online electronic health record(EHR) enables patients to centrally manage the own medical records, which

greatly facilitates the storage, access and sharing of personal health data. With the emergence of cloud com-

puting, it has succeeded in attracting attention and transferring their EHR applications to an efﬁcient system

for storing and accessing data. However, due to lose physically control of personal data in a cloud computing

circumstance, it brings about a serious privacy problem for the data owner. Therefore, cryptography schemes

offering a more suitable solution for enforcing access policies based on user attributes are needed. We have

proposed a framework with ﬁne-grained access control mechanism that protects electronic health data in va-

rieties of devices, including smart mobile device. We make EHR security through designing online/ofﬂine

key policy attribute-based encryption scheme which is an extension of identify-based encryption (IBE). This

scheme can provide ﬁne-grain access policy and efﬁciency for users’ data. Especially, it’s greatly reducing

complexity and computational of encryption and key generation.

1 INTRODUCTION

Alone with progressively advanced of technology de-

velopment, it brings more convenient to the whole so-

ciety. At present, people is available to use majority

kinds of portable equipment, which equip full acces-

sing control performance, to administer their body’s

medical and health information. Everyone is avai-

lable to observe themselves health records to great

understand body situations bypassing their common

action and exercise training, conveniently catching

whether health maintains good conditions or not at

any time. Technically, a personal health device has

to necessitate the interaction of health records, as the

data results are cared about by user with analyzing

in the storage center. Consequently, those services

permit a person to generate, manage and manipulate

health records through the Internet anywhere and at

any time. The feedback of analyzing reports plays the

friend and beneﬁt role in the process between users

and data analysis center. At the same time, users can

have overall control over their health records and be

able to effectively compare their health data with ot-

her health care providers, not just their families and

friends. In this way, physicians can take advantage of

different physiological parameters to make diagnosis

more accurate and efﬁcient, while reducing the pa-

tient’s medical expenses.

However, the EHR system must have charge of

the burden of storage, computing and communica-

tion pressures from a growing number of user. There-

fore, its services must have powerful computing abi-

lity. Nowadays, a new computing paradigm called the

cloud has possession of the dynamic extension ability,

which ﬂexibly connects to the network by obtaining

computing resources and services. For the reason of

its virtually unlimited and ﬂexible storage and com-

puting ability (Fan et al., 2015), the cloud computing

has attracted much concern to research and apply in

academic and industry. It is obvious that the cloud

holds gigantic saved space and application ability to

turn into preferred data processing rather than running

in the specialized data center. Thus, the EHR system

is imperative to outsource its data and computations to

cloud for storage and calculation. In addition, it also

can heavily reduce the consumption of EHR opera-

tion. Likewise, an increasing number of patients and

physician have transformed their habits to visit EHR

system in smart mobile equipment, which is easier

access and operation than traditional methods. We

can take some applications of smart devices as an

example, such as MediTouch EHR Software, Ofﬁce

110

Liu, K.

Secure Electronic Health Record System Based on Online/Ofﬂine KP-ABE in the Cloud.

DOI: 10.5220/0006350101100116

In Proceedings of the 2nd International Conference on Internet of Things, Big Data and Security (IoTBDS 2017), pages 110-116

ISBN: 978-989-758-245-5

Practicum from mobile application providers.

Although we enjoy utilizing the convenience and

intellect from EHR service in the cloud, our privacy

is undergoing unprecedented menace in the process

of information processing and communication. The

EHR system and storage service have possessed a

lot of sensitive information about user privacy, which

is greatly anxious in data providers. That is to say,

the user’s sensitive data is in potentially dangerous

circumstance which they don’t know whether EHR

would be leakage in the EMR system. These are tre-

mendous security concerns that users have no way to

trust its service and fully rely on the EMR system.

That is to say, when the data is reserved in the cloud

server, the most major anxiety of users is about the

secret key management and authority issues because

they have no idea who can be authorized to visit the

data in the EHR system. At the same time, ever since

the patients uploaded and outsourced their record into

the cloud server, they have lost actual control about

their medical and health data in physical. Just be-

cause of this, those important information could have

no longer to be safely guaranteed full privacy assure.

On the one hand, due to misbehave from the cloud in-

ner servers, the patients’ private data may leak to the

outside. Such error behavior may also lead to reveal

the worthy information in personal data, including te-

lephone numbers and medical records. For instance,

The malicious adversary can utilize those personal in-

formation to recommend special medicine for seeking

illegitimate interests. On the other hand, cloud ser-

vers are more vulnerable suffering malicious attacks

from outside, which is decided by its unique features

that cloud servers calculus and store data on the open

platform.

To address the potential risks of privacy exposu-

res, EHR system needs to construct a kind of access

control mechanism that permit patients to adminis-

ter the entire control to do the selectively share their

own EHR information rather than making the cloud

encrypted the patients’ data. In this respect, the

access control of EHR should be provided in tradi-

tional way, apart from secured by the server (Benaloh

et al., 2009). Generally, any decryption key should

be generated by corresponding patient, which is avai-

lable to be distributed for authorized users. There-

fore, data owner should be granted permission to visit

and revoke their record which possesses crucial effect

to guarding data owner’s privilege to their sensitive

information while preserving their privacy (Mandl

et al., 2001). To resolve the ﬁne-grained access cont-

rol and sharing issues, Sahai and Waters put forward

a notion of attribute-based encryption (ABE), derived

from a type of application Fuzzy-IBE scheme (Sa-

hai and Waters, 2005). ABE targets to offer ﬁne-

grained and extensible access control to encrypt plain-

text, which is according to a kind of one to multiple

public key encryption schema attracting lots of atten-

tion to research. However, it is a signiﬁcant drawback

that the computational complexity of access control

rule and account of attributes directly make a diffe-

rence of efﬁciency of the encryption and the key ge-

neration phases. In protecting privacy of EHR system,

these issues bring obstacles to preserving the privacy

of EMR system, which leads to increase consumption

in plaintext encryption phase and user key generation

phase (Hohenberger and Waters, 2014).

This paper aims to eliminate this security and pri-

vacy problem through constructing online/ofﬂine key-

policy attribute-based encryption scheme (OO-KP-

ABE). By comparing with the previous KP-ABE, we

split the encryption into different phases, online en-

cryption and ofﬂine encryption, which enable to de-

crease the complexity and computation of encryption

and key generation for data owner.

The remainder of this paper is arranged as follows.

We recall the related work in section 2, including the

development of EHR and OO-KP-ABE scheme. Furt-

hermore, section 3 illustrates some preliminary nation

about bilinear group, access control structure and on-

line/ofﬂine key-policy attribute. In section 4, we pre-

sent our main framework of EHR cloud system inclu-

ding security requirement and analysis. Finally, this

paper is summarized in section 5.

2 RELATED WORK

In order to improve patients care, safety and costs sa-

ving, health record system is undertaking to achieve

modernization for greater efﬁciency with advance

side by side (Buck, 2007; Kim et al., 2008; Lohr,

2009; Tripathi et al., 2009; Health et al., 2008). Com-

monly, an EHR system contains a digital record about

patients’ fundamental medical and health data that is

available to provide for authorized clinicians and staff

to create, concentrate, manage and view. In general,

Hospitals or research institutions usually have pos-

session and store only one aspect of the patient’s re-

cord. Patients and doctors are able to exert repeatedly

checking and incurring unnecessary costs in EHR sy-

stem.

We have implemented IBE scheme to encrypt data

of a single user in EHR system, after which data ow-

ner must know the user’s identity until send the en-

crypted data to the user. Previously, We constructed

an IBE scheme for individual users to encrypt data.

In other words, before data owner dispatches his en-

Secure Electronic Health Record System Based on Online/Ofﬂine KP-ABE in the Cloud

111

crypted personal data to user, he must know and con-

ﬁrm the identity (Boneh and Franklin, 2001). Actu-

ally, this IBE operation would not take effect when

the sending part didn’t verify the deﬁnite the user’s

identity. To solve the problem above, ABE is avai-

lable to adapt to the one to many encryption scena-

rios. Especially, there has a tremendous tendency to

use ABE for personal health record (PHR). Conse-

quently, PHR system must satisfy ﬁne-grained access

control strategy and make efﬁcient revocation possi-

ble, while KP-ABE can be achieve those requirements

(Yan et al., 2016; Liu et al., 2016; Liu et al., 2016; Li

et al., 2015). But the ciphertext length linearly incre-

ase the account of users, which also brings a heavy

burden on the computational complexity and compu-

tation of the server.

The conception of attribute-based encryption was

derived from fuzzy IBE by Sahai and Waters in (Sahai

and Waters, 2005), which was coped with by Goyal

et al at ﬁrst. Then, ABE are available to allow data

owner to transmit ciphertext to users based on cer-

tainly speciﬁed access policies. Simultaneously, there

are categorized into two kinds of ABE scheme, key-

policy attribute-based encryption (KP-ABE) (Goyal

et al., 2006) and ciphertext-policy attribute-based en-

cryption (CP-ABE) (Bethencourt et al., 2007). When

correspondent ciphertext is satisﬁed the access cont-

rol policy connected with a speciﬁc set of attribute,

the private key can be generated accordingly. Na-

mely, the secret key is connected to access control

structure. The user can perform the decryption algo-

rithm to obtain the plaintext through a corresponding

private key whose attribute set is exclusively authori-

zed a set of the private key?s access control structure.

KP-ABE is a cryptography system based on bilinear

map and Linear Secret Sharing Schemes (LSSS). On

the other hand, the ciphertext relates the access cont-

rol policy (Bethencourt et al., 2007) in CP-ABE.

Although ABE has possession of powerful

function, there is a concern about efﬁciency drawback

to conﬁne it practical application. In most ABE sy-

stem, the encryption and decryption costs grow with

the account of attributes and the complexity of access

structure. In particular, mobile devices must run en-

cryption and decryption algorithms to protect their

real-time data. The required time to run and cache

may give rise to sustain problems for limiting battery

power supply. In 2014, for sake of alleviating the cus-

tom of mobile instruments, Hohenberger et al. (Rou-

selakis and Waters, 2013) has proposed methods for

online/ofﬂine encryption in ABE setting.

3 PRELIMINARIES

3.1 Bilinear Mapping

Deﬁnition 1 (Bilinear mapping (Boneh et al.,

2004)) Let P

, P

be two multiplicative cyclic groups

of prime order p. Let g be the generator of P

. Deﬁne

a bilinear map e : P

× P

→ P

. It has the following

properties:

• Bilinear: e(g

, g

) = e(g

, g

)

for all a, b ∈ Z

and g

, g

∈ P

• Non-degenerate: e(g, g) 6= 1.

If the group operation in P

and the bilinear map e are

both computable, the multiplicative cyclic group P

a bilinear group. It is a remarkable fact that the map e

is symmetric since e(g

, g

) = e(g

, g

)

= e(g

, g

3.2 Access Structure

Deﬁnition 2 (Access Structure (Beimel, 1996)) Let

U = {U

, U

, ··· , U

} be a set of parties. A col-

lection A ⊆ 2

,···,U

}

is monotone if ∀B, C, if

B ∈ A and B ⊆ C denote C ∈ A. An access structure

is a monotone collection A of non-empty subsets of

, U

, ··· , U

} (that is A ⊆ 2

,···,U

}

\ φ). The

sets in A are called the authorized sets, and the sets

not in A are called the unauthorized sets.

The access structure A plays a vital role in authori-

zing sets of attribute in our context. We constrict it as

monotone access structures, formally.

3.3 Online/Ofﬂine KP-ABE Scheme

We illustrate the KP-ABE schema with online/ofﬂine

encryption (Hohenberger and Waters, 2014) by the

following ﬁve polynomial algorithms.

• Setup(λ, U ): A security parameter λ and a uni-

verse U of attributes take account into the setup

algorithm. We choose a bilinear group P of prime

order p ∈ Θ(2

).It needs stochastically choosing

generators g, h, u, w ∈ P and picking a random ex-

ponent α ∈ Z

. Then, it sets up the keys as:

PK = (P, p, g, h, u, w, e(g, g)

), MK = (PK, α).

Firstly, we suppose that the university of attributes

can be encoded as elements in Z

• Extract(MK, (M, ρ)): The extract algorithm takes

the master secret key MK and a Linear Serect

Sharing Scheme access structure (M, ρ) from a

user as an input. The trusted key generation cen-

ter provides a corresponding private key SK

(M,ρ)

to give the user by working extract algorithm.

IoTBDS 2017 - 2nd International Conference on Internet of Things, Big Data and Security

112

• Ofﬂine Encryption(PK): The ofﬂine encryption

algorithm just needs the encrypter to take the

public parameters PK as input. The encrypter

will engender and store an intermediate ciphertext

pool IT .

• Online Encryption(PK, IT, S): During the online

encryption phase, the encrypter is available to out-

put a ciphertext CT and produce a session key key

which is kept to itself before inputting the public

parameters PK , an intermediate ciphertext pool

IT , and a set of attributes S.

• Decryption(PK, CT, SK

(M,ρ)

): A user utilizes the

public parameters PK , its private key SK

(M,ρ)

for

access structure (M, ρ) and a ciphertext CT asso-

ciated with an attribute set S as input. It is decap-

sulated CT to recover a session key key or inputs

the distinguished symbol ⊥.

The correctness condition as well as the model for

deﬁning the adaptive security for online/ofﬂine KP-

ABE is provided in (Hohenberger and Waters, 2014).

3.4 Security Model of OO-AB-Key

Encapsulation Mechanism

Based on hading a symmetric session key, the

key encapsulation mechanism (KEM), which is able

to use in the online/ofﬂine KP-ABE, can be uti-

lized for encrypting data with arbitrary length at

the symmetrical encryption scheme. The follo-

wing selective-set model for online/ofﬂine KP-ABE

scheme was given by Hohenberger et al. (Ho-

henberger and Waters, 2014), which illustrates the

game as follows. Initiatorily, we deﬁne that Π =

(Setup, Extract, O f f .Enc, On.Enc, Decryption) is an

AB − KEM (Rouselakis and Waters, 2013) for access

structure space P, and consider the below game for

parameter λ, attribute universe U and an adversary A .

• Setup: The public parameter PK is generated in

the process of a challenger running the setup al-

gorithm. Then he sends it to the adversary.

• Step 1: It is necessary to initialize an empty table

T , an empty set D and an integer counter j = 0,

which is performed by the challenger. Adaptively,

the adversary is able to arbitrarily perform the fol-

lowing inquiry:

1. Create (I

key

): The challenger, who require to

run the key generation algorithm on I

key

after

setting j := j + 1, acquires the secret key SK

and stores the entry ( j, I

key

, SK) in the table T .

It is worthy noted that Create(·) operation could

be inquired again and again with the same in-

put.

2. Corrupt (i): If there consists of an i

entry

in table T , after that the challenger is able to

acquire the entry ( j, I

key

, SK) and sets D := D

and I

key

. It is the next to return the private key

of adversary SK. Then, it would return ⊥, if no

such entry exists.

3. Decrypt (i, CT ): When table T has an i

en-

try in existence, the challenger could obtain the

entry the entry ( j, I

key

, SK). Accordingly, he

can return to the adversary the output of the de-

cryption algorithm on input (SK, CT ). Then, it

would return ⊥, if no such entry exists.

4. Challenge: In order to make all I

enc

∈ D,

f (I

key

, I

∗

enc

) 6= 1, the provides a challenge value

∗

enc

ﬁrstly. The challenger acquires (I

key

, I

∗

enc

)

bypass running the algorithm Online.Encrypt

(PK, O f f line.Encrypt(PK), I

∗

key

). After that,

it chooses a bit b at random. If b = 1, it se-

lects a random session key R in the session key

space and returns (R, CT

∗

). If b = 0, it returns

(key

∗

, CT

∗

) to the adversary.

• Step 2: Step 1 is associated with the constrictions

that the adversary aren’t able to trivially acquire a

private key for the challenge ciphertext. Namely,

it satisﬁes f (I

key

, I

∗

enc

) = 1 being added to D. And

it cannot also send a decryption query about the

challenge ciphertext CT

∗

• Guess: While the adversary outputs a guess b

b, the output of the experiment is 1 if and only if

b = b

Deﬁnition 3 (Online/Ofﬂine AB-KEM Security)

For all probabilistic polynomial-time adversaries

A, there is attribute universe U if under chosen-

ciphertext attack (CPA) for the semantic security,

there exists a negligible function negl such that:

Pr[OOAB − KEM

A,Π

(λ, U) = 1] 6

+ negl(λ)

CPA Security. If we take away the Decrypt oracle in

both step 1 and 2, we view that a system satisﬁes CPA-

secure (or secure against chosen-plaintext attacks).

4 THE FRAMEWORK OF EHR

COULD SYSTEM

4.1 EHR System Illustration

The interaction between participants and EHR cloud

system is precisely illustrated through the ﬁgure 1. It

is included the health data producer, the data user and

the trusted key generation key center and the cloud

Secure Electronic Health Record System Based on Online/Ofﬂine KP-ABE in the Cloud

113

EMR Cloud

Send request with attribute

Offline

encryption

phase

Data Owner

Online

encryption

phase

EMR User

Trusted Center

Ciphertext

Intermediate

Ciphertext

Public Key

Master Key

3. Private Key (if pass)

1.Setup

Policy Decision

2. Encryption

Download

5.Decryption:

Message

Figure 1: System CP-ABE and workﬂow.

service. The framework illustrates how the data ow-

ners store their personal data.

• EHR system setup: EHR system running the se-

tup algorithm requires that the trusted private key

generation center chooses a security parameter λ

to product public key PK and master secret key

MK by a random oracle model. Then, the trus-

ted center is able to transmit the public key PK to

the data owner. Consequently, the trusted center

holds the master key MK to itself.

• Encryption: The traditional encryption phase is

divided into two sessions. On the ofﬂine encryp-

tion phase, the data owner runs the ofﬂine encryp-

tion algorithm to generate intermediate ciphertext

IT . Especially, IT must be store securely by the

user in the local side. On the online encryption

phase, the data owner has to run the online encryp-

tion algorithm through taking as the message M,

the public parameter PK, the assess control struc-

ture (M, ρ) and the set of attribute U into input.

After that, the user is able to outsource their cip-

hertext into the cloud.

• The request: The authorized request is sent to the

trusted center by the data user, who takes as his

attributes into input. If the request would be veri-

ﬁed into the truth, the trusted center could give a

?pass? answer to the data user. If answers were

negative, the trusted center would refuse this re-

quest from the data user.

• Key Extract: If the attribute of user satisﬁes the

access control assess control structure (M, ρ), the

trusted center would be obtained a ”pass” ans-

wers. As a consequence, the trusted center will

run Extract algorithm of OO-KP-ABE schema to

gain the private SK. Then, it sends SK to the user

to decrypt the ciphertext CT .

• Decryption phase: When the data user obtains the

positive answer, the trusted center immediately

transforms the private key to him. With taking as

privacy key and the ciphertext CT from the cloud

into input, the data user is available to compute

the message through running the decryption algo-

rithm of online/ofﬂine KP-ABE.

The data user sends an authorized request which

the mobile device data producer and medical institu-

tion, for the beginning, the trusted center utilize the

Extract algorithm to generate a privacy key SK, which

takes parameter key PK and master key MK as input.

4.2 Security Requirements

In the following, We introduce four characteristics of

the main security requirements in the EHR cloud sy-

stem.

• Data Conﬁdentiality: When electronic health re-

cord of data owner stores in the cloud, it couldn’t

be arbitrarily consulted if someone doesn’t own

private key and possess patients’ authorization.

Thus, the EHR has to be encrypted before uplo-

ading to the cloud server. Owing to encrypt data

though patients’ attribute, only someone obtains

the private key who have been authorized.

• Authenticity: Data owner reserves their informa-

tion in the EHR system, which implements varie-

ties of operation in the third cloud platform. Pa-

tients don’t permit their valuable and sensitive re-

cord to be compromised and distorted by mali-

cious attackers. Hence, The cloud server has to

guarantee the authenticity of data for data provi-

der. Similarity, we can ensure the authenticity of

data for verifying the information and improving

access control in especial scheme, such as (Meng

et al., 2016; Yan et al., 2016), respectively. For

another approach, the user accesses data used to

study or treat other patients. If the patient’s EHR

is not reliable, this will be a big hidden dangers.

• Privacy protection for patients: The number of

access control population should be conﬁned by

the purpose of the visitor. According to the user’s

attribute differences, he has different access per-

missions. For instance, doctor, nurse and resear-

cher, they have possession of authority to access

data based on their usage purpose in the EHR sy-

stem.

• Revocation: If a user want to revoke his attribute,

after that immediately the user will not be avai-

lable to access EHR using that attribute, known

as attribute revocation. There is another situation

users can no longer use corresponding private key

IoTBDS 2017 - 2nd International Conference on Internet of Things, Big Data and Security

114

when the data owner makes a time limit for his

one of access control.

5 CONCLUSIONS

In this paper, we construct a scheme for uncertain

users and a ﬁne-grained access control of EHR system

by attribute policy in the cloud. Traditional public

key encryption system is unsuitable to encrypt mul-

tiple to one or multiple to multiple situation. Previ-

ously, access control is aimed to a single known user

who only delegates a known identity. Nowadays, pe-

ople are available to record their health data in EHR

system by moving electronic devices. This function

isn’t limited by time and place, which only needs de-

vice having sufﬁcient power and communication In-

ternet. Consequently, a semi-trusted third cloud plat-

form provides these service in our schema. Moreover,

the patient must have complete control power over

their own data, such as specifying a particular person

viewing the data set, and those who do not match the

attribute policy do not have access to the data set.

As above, the OO-KP-ABE scheme is that the pro-

mising technology should speed up the utilization of

EHR cloud platform in other related works in an elec-

tronic health ﬁeld.

ACKNOWLEDGEMENTS

This work was supported by National Natural Science

Foundation of China (No. 61472091), Natural

Science Foundation of Guangdong Province for Dis-

tinguished Young Scholars (2014A030306020) and

Science and Technology Planning Project of Guang-

dong Province, China (2015B010129015).

REFERENCES

Beimel, A. a. (1996). Secure schemes for secret sharing and

key distribution. Technion-Israel Institute of techno-

logy, Faculty of computer science.

Benaloh, J., Chase, M., Horvitz, E., and Lauter, K. (2009).

Patient controlled encryption: ensuring privacy of

electronic medical records. In Proceedings of the 2009

ACM workshop on Cloud computing security, pages

103–114. ACM.

Bethencourt, J., Sahai, A., and Waters, B. (2007).

Ciphertext-policy attribute-based encryption. In Secu-

rity and Privacy, 2007. SP’07. IEEE Symposium on,

pages 321–334. IEEE.

Boneh, D., Di Crescenzo, G., Ostrovsky, R., and Persi-

ano, G. (2004). Public key encryption with keyword

search. In International Conference on the Theory

and Applications of Cryptographic Techniques, pages

506–522. Springer.

Boneh, D. and Franklin, M. (2001). Identity-based encryp-

tion from the weil pairing. In Annual International

Cryptology Conference, pages 213–229. Springer.

Buck, C. F. (2007). Designing a consumer-centered perso-

nal health record. Technical report, Technical report,

California Health Foundation.

Fan, K., Huang, N., Wang, Y., Li, H., and Yang, Y. (2015).

Secure and efﬁcient personal health record scheme

using attribute-based encryption. In Cyber Security

and Cloud Computing (CSCloud), 2015 IEEE 2nd In-

ternational Conference on, pages 111–114. IEEE.

Goyal, V., Pandey, O., Sahai, A., and Waters, B. (2006).

Attribute-based encryption for ﬁne-grained access

control of encrypted data. In Proceedings of the 13th

ACM conference on Computer and communications

security, pages 89–98. Acm.

Health, U. D., Services, H., et al. (2008). The nation-

wide privacy and security framework for electronic

exchange of individually identiﬁable health informa-

tion. Ofﬁce of the National Coordinator for Health

Information Technology.

Hohenberger, S. and Waters, B. (2014). Online/ofﬂine

attribute-based encryption. In International Workshop

on Public Key Cryptography, pages 293–310. Sprin-

ger.

Kim, G. R., Lehmann, C. U., on Clinical Informa-

tion Technology, C., et al. (2008). Pediatric aspects of

inpatient health information technology systems. Pe-

diatrics, 122(6):e1287–e1296.

Li, J., Li, J., Chen, X., Jia, C., and Lou, W. (2015).

Identity-based encryption with outsourced revocation

in cloud computing. Ieee Transactions on computers,

64(2):425–437.

Liu, J. K., Au, M. H., Huang, X., Lu, R., and Li, J. (2016).

Fine-grained two-factor access control for web-based

cloud computing services. IEEE Transactions on In-

formation Forensics and Security, 11(3):484–497.

Lohr, S. (2009). Ge and intel join forces on health techno-

logies. New York Times, 3.

Mandl, K. D., Markwell, D., MacDonald, R., Szolovits, P.,

and Kohane, I. S. (2001). Public standards and pa-

tients’ control: how to keep electronic medical records

accessible but privatemedical information: access and

privacydoctrines for developing electronic medical re-

cordsdesirable characteristics of electronic medical

recordschallenges and limitations for electronic medi-

cal recordsconclusionscommentary: Open approaches

to electronic patient recordscommentary: A patient’s

viewpoint. Bmj, 322(7281):283–287.

Meng, D., Wang, W., Luo, E., and Wang, G. (2016).

Attribute-based traceable anonymous proxy signature

strategy for mobile healthcare. In Security, Privacy,

and Anonymity in Computation, Communication, and

Storage: 9th International Conference, SpaCCS 2016,

Zhangjiajie, China, November 16-18, 2016, Procee-

dings 9, pages 178–189. Springer.

Rouselakis, Y. and Waters, B. (2013). Practical con-

structions and new proof methods for large universe

Secure Electronic Health Record System Based on Online/Ofﬂine KP-ABE in the Cloud

115

attribute-based encryption. In Proceedings of the 2013

ACM SIGSAC conference on Computer & communi-

cations security, pages 463–474. ACM.

Sahai, A. and Waters, B. (2005). Fuzzy identity-based

encryption. In Annual International Conference on

the Theory and Applications of Cryptographic Techni-

ques, pages 457–473. Springer.

Tripathi, M., Delano, D., Lund, B., and Rudolph, L. (2009).

Engaging patients for health information exchange.

Health Affairs, 28(2):435–443.

Yan, H., Li, J., Li, X., Zhao, G., Lee, S.-Y., and Shen, J.

(2016). Secure access control of e-health system with

attribute-based encryption. Intelligent Automation &

Soft Computing, 22(3):345–352.

IoTBDS 2017 - 2nd International Conference on Internet of Things, Big Data and Security

116