Healthcare Provision in the Cloud: An EHR Object Store-based Cloud

Used for Emergency

Chrysostomos Symvoulidis

1,2

, Athanasios Kiourtis

, Argyro Mavrogiorgou

and

Dimosthenis Kyriazis

Department of Digital Systems, University of Piraeus, Piraeus, Greece

BYTE S.A., Athens, Greece

Keywords:

EHR Cloud, Object Storage, Emergency, Storage Clouds.

Abstract:

EHR Cloud architectures have come a long way within the last years, giving the ability to individuals to

store their healthcare data in the cloud, thus being accessible at all times. Though Electronic Health Records

(EHR) and Personal Health Records (PHR) sharing technologies have been developed over the last decades,

and a lot of the attention is given on the exchange of healthcare data between organizations and healthcare

institutions, less emphasis has been put in the services regarding the exchange of such data between individuals

and healthcare professionals and the issues that this gap creates are yet to be answered. To this end, in this

paper, we introduce an EHR Cloud-based system that utilizes an Object Storage architecture to store healthcare

data, and provides the ability to authenticated healthcare professionals to gain access if needed during an

emergency, in an automated yet secure way, for accelerated health services provision. The proposed approach

is evaluated and the results are presented in order to justify the rationale behind its design.

1 INTRODUCTION

Cloud computing has come a long way during the

last two decades, becoming a state of the art solution

for the provision of services across all areas; from ﬁ-

nance and banking systems, to hosting the elections

of a country. Let alone, the healthcare industry where

the cloud can play a huge role.

The adoption of cloud in the healthcare leads to

great changes that impact highly the way the health-

care professionals work (Vinati Kamani, 2019), mak-

ing their job easier, faster, and more efﬁcient. For

instance, there has been a major impact on the way

the produced health-related data are managed and an-

alyzed. Healthcare data regard a signiﬁcant asset that

without cloud computing could not be exploited in

its full potential. Powerful cloud-hosted services that

utilize techniques like Big Data analytics and Artiﬁ-

cial Intelligence (AI), can now analyze tons of new-

coming data at a glance, making the process not only

easier, but faster too, while expanding several possi-

bilities (Martin, N., 2019; Davenport and Kalakota,

2019).

Interoperability issues over healthcare systems

across different organizations regards a great concern

that until now is not entirely dealt with. But, big steps

are taken towards that direction, and cloud comput-

ing assists considerably. A typical example, regards

the report by West Monroe Partners (Cohen, 2018),

stating that 35% of the questioned organizations store

more than 50% of their healthcare data in the cloud.

This brings new potential to solving the interoper-

ability issues, by allowing inter-institutional data ex-

change.

Nevertheless, the impact that cloud computing

leaves on the healthcare sector is mostly related on

evolving the systems and procedures in healthcare in-

stitutions, something crucial and of high importance.

But small, or rather less attention is given on the cre-

ation of services that could give the ability to individ-

uals to collect and manage their own healthcare data,

as well as exchanging their data with professionals

when needed.

As indicated in Section 2 as well, there exist so-

lutions that complement such requirements but their

main focus is concentrated on the security aspects

rather than trying to agile and speed up the whole

data exchange process. For this reason, in this paper

we propose a novel EHR Cloud architecture that al-

lows its users to safely store their healthcare data,and

more precisely their Electronic Health Record (EHR)

in the Cloud, while giving the possibility to healthcare

professionals to gain access to this data when needed,

without supplanting the serious security and privacy

Symvoulidis, C., Kiourtis, A., Mavrogiorgou, A. and Kyriazis, D.

Healthcare Provision in the Cloud: An EHR Object Store-based Cloud Used for Emergency.

DOI: 10.5220/0010247004350442

In Proceedings of the 14th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2021) - Volume 5: HEALTHINF, pages 435-442

ISBN: 978-989-758-490-9

435

issues that should also be taken under consideration.

The remaining of the paper is constructed as fol-

lows. Section 2 provides a description regarding the

study of the state of the art in the ﬁelds of EHR

Storage Cloud and Secure EHR Management in the

Cloud. Section 3 provides a detailed description of an

emergency scenario where the proposed EHR Cloud

architecture can be exploited in its full potential. Sec-

tion 4 describes the EHR Cloud architecture in de-

tail. Section 5 analyzes the experiments performed

in order to evaluate the proposed EHR Cloud. In Sec-

tion 6, the requirements of an EHR-related Cloud with

respect to security and privacy are introduced, along

with the way these are fulﬁlled by the proposed EHR

Cloud. Finally, Section 7 concludes the paper.

2 RELATED WORK

This section summarizes the work achieved in the

area of EHR Storage Clouds and EHR Management

in the Cloud. There exist several storage cloud ser-

vices solutions for the secure collection of EHRs and

healthcare data in general. In (Cai et al., 2017), the

authors propose an EHR sharing scheme deployed in

the cloud where the owner of the healthcare data can

generate EHR ciphertexts, for a user to be able to de-

crypt them based on transformed ciphertexts from the

EHR Cloud.

Moreover, the authors in (Cao et al., 2019) pro-

pose a secure cloud-assisted e-Health system, where

the medical institutions that generate and own a pa-

tient’s EHR ensure that only authorized individuals

may have access and be able to modify those EHR

data, without the involvement of a trusted entity.

The latter is achieved with the exploitation of the

blockchain technology, providing a tamper-prooﬁng

way to operate actions including the exchange of

EHRs, since no transaction can be accepted unless

documented into the blockchain.

In a relevant manner, in (Seol et al., 2018) a cloud-

based EHR model, utilizing Attribute-based access

control techniques is proposed. In this research, the

main focus is given on securing the transactions be-

tween the owner of the EHR and its requestor. All

transactions are digitally signed prior to their exe-

cution, while partial encryption of the EHRs is per-

formed as well. In addition, in (Joshi et al., 2018), a

centralized attribute-based authorization mechanism

is introduced. In this study, Attribute Based Encryp-

tion is used allowing medical organizations to del-

egate authority to healthcare providers in accessing

EHRs stored in cloud-based systems. Finally, the au-

thors of (Manoj et al., 2017) suggest the use of two en-

cryption methods in a hybrid EHR system, aiming at

achieving protection of data privacy and access con-

trol upon the health data.

Based on the research work described above, it is

made clear that a lot of work has been achieved in the

area of secure transaction of EHRs and health data be-

tween two or more entities and storage of such data in

Cloud-based infrastructures. What can be identiﬁed

based on the above-mentioned research work it that

the attention is mainly focused on the security and

privacy risks that arise when the exchange of health

data between healthcare organizations. However less

attention has been given on how EHR exchange can

be achieved between an individual and healthcare or-

ganizations, with respect to the efﬁciency of such sys-

tems.

With that in mind, in this paper we introduce a

novel EHR Cloud system utilizing the Object Stor-

age architecture (Factor et al., 2005) whose purpose

will be two-fold. To give the user the ability to safely

store and backup their healthcare data in the Cloud,

and allow Healthcare Professionals to gain access to

this data if needed in an automated yet secure way, for

accelerated health services provision.

3 EMERGENCY SCENARIO

This section describes the scenario that worked as a

thriving force in order to design the EHR Cloud ar-

chitecture which will be described in section 4.

3.1 Preliminaries

In order to better comprehend the scenario it is very

important to deﬁne the entities that constitute the

overall system.

• Electronic Health Record Application (EHR

App): A smartphone application that is used by

the user. Through this application a user is able

to access their EHR that is stored locally on the

phone and visualized through this application. In

addition, using the same application the user up-

loads their health data (e.g. EHR, Medical Im-

ages, etc.) to the EHR Cloud.

• Healthcare Professional Application (HCP

App): The Healthcare Professional’s (HCP) ap-

plication through which an HCP gains access to

the user’s health data that is stored in the EHR

Cloud. Using the same application, the HCP may

also visualize this data, modify it and upload it

to the EHR Cloud. The HCP App in the current

HEALTHINF 2021 - 14th International Conference on Health Informatics

436

Figure 1: Emergency scenario.

emergency scenario replicates the system used in

Hospitals and other Healthcare institutions.

• Health Record Index (HRI): A health record

indexing methodology as proposed by (Kiourtis

et al., 2020) through which HCPs may access

a user’s EHR that is stored in a Storage Cloud

similar to the proposed EHR Cloud in emer-

gency cases. In cases where several EHR Cloud

providers exist, the HRI holds information about

each user’s preferred EHR Cloud.

• EHR Cloud: The proposed EHR Cloud architec-

ture as proposed in Section 4. A user utilizing the

EHR App on their phone uploads their health data

in the EHR Cloud.

• Certiﬁcation Authority: An entity that can cer-

tify an individual as HCP. This is a mandatory en-

tity in order to assure that only certiﬁed individ-

uals can gain access to the EHR Cloud in emer-

gency situations.

3.2 Scenario Description

The emergency scenario, also visually depicted in

Figure 1, is split into two time periods. The ﬁrst in-

cludes the steps prior to the emergency, while the sec-

ond starts at the moment the emergency occurs and

can be further divided into the following steps.

1. Pre-emergency:

(a) Upload EHR: The ﬁrst step of this emergency

regards the upload of the user’s EHR to the

EHR Cloud, through the EHR App. An impor-

tant note, is that the EHR is encrypted on the

smartphone side prior to its upload to the EHR

Cloud.

(b) QR-code Creation: As soon as the upload is

complete, a QR-code is created in the smart-

phone application. This QR-code contains the

information needed by the HCP in order to con-

tact the HRI, gain access to the user’s EHR

that is stored in the EHR Cloud and decrypt the

EHR. This QR-code is printed by the user and

kept with them at all times.

2. An Emergency Occurs:

(a) User in Medical Need: An EHR Cloud user is

in need for medical assistance and is transferred

to the nearest medical centre.

(b) QR-code Scan: The HCP collects the above-

mentioned QR-code from the user and scans it

through the HCP application.

code holds information about the citizen. This

information is queried to the HRI service which

returns the EHR Cloud that the user used to up-

load their EHR.

Healthcare Provision in the Cloud: An EHR Object Store-based Cloud Used for Emergency

437

(d) Download EHR Request: The HCP requests

to download the EHR from the user’s preferred

EHR Cloud. This cannot be achieved unless

the HCP is already certiﬁed by a Certiﬁcation

Authority.

(e) HCP Certiﬁcation: The HCP provides the

EHR Cloud the necessary information in order

to certify themselves as HCPs. The EHR Cloud

then contacts this service in order to verify the

identify of the HCP. If the process of certifying

is unsuccessful, the request to grant access to

the EHR is aborted.

(f) Download EHR to the HCP App: This steps

can only happen, when an HCP is in fact certi-

ﬁed as HCP. If the HCP is certiﬁed, the down-

load of the EHR begins.

(g) EHR Decryption: The EHR is now down-

loaded in the HCP App. Using the above-

mentioned information collected from the QR-

code the EHR is decrypted and visualized.

4 CLOUD-BASED EHR

ARCHITECTURE

This section describes the proposed EHR Cloud ar-

chitecture. In Figure 2 the proposed EHR Cloud’s

architecture is presented. There, four main compo-

nents are identiﬁed. The (i) Object Store where the

users’ encrypted EHR are stored, (ii) the Identity

manager whose purpose is two-fold; to hold infor-

mation related to the user’s account and create tem-

porary accounts for the HCPs in order to access the

EHR Cloud. The remaining components of the pro-

posed EHR Cloud are (iii) the Access Auditing com-

ponent that keeps records of who and when accessed

the EHR Cloud along with the ﬁles that were added,

deleted or modiﬁed, while (iv) the HCP Certiﬁcation

checking mechanism ensures that only someone cer-

tiﬁed as HCP by a trusted certiﬁcation authority HCP

can access the EHR Cloud in case of an emergency.

Note that for simplicity reasons the HR Index and the

acknowledgement messages after each action are not

shown in Figure 2.

4.1 The Object Store

The primary component of the proposed architec-

ture is the Object Store where the users’ EHRs are

kept. The implemented EHR Cloud solution exploits

MinIO (MinIO Inc., 2020), a high performance ob-

ject storage developed especially for building cloud-

native applications and service, as the EHR Cloud.

Not only that, an object storage is preferred, instead

of a NoSQL-based architecture that is commonly used

for similar projects (Sreekanth et al., 2015) for the fol-

lowing reasons:

• The ﬁles stored in an object storage are by deﬁ-

nition as objects. These objects contain the data

along with its metadata and a unique identiﬁer,

thus making object stores highly customizable

and powerful.

• Object stores can be deployed in commodity, less

expensive hardware making them easier and less

costly to manage and upgrade.

• Scalability is relatively easy to accomplish as

well, something crucial for systems that require

efﬁciency and low response time, as the proposed

EHR Cloud.

• Object stores also ensure high availability for the

stored data. For this reason, storing unstructured

data, photos and videos (e.g. Medical Images,

etc.) regard an optimal use case for using Object

stores.

The objects (i.e. the user’s EHR) stored in the

EHR Cloud are encrypted prior their upload to the

EHR Cloud on the phone side. This policy is set in

order to ensure that even if in the worst-case scenario

where an unauthorized entity manages to access the

EHR Cloud, they will not be able to gain access to the

health record itself.

4.2 The Identity Manager

The Identity Manager is another crucial component

of the proposed architecture, since it is responsible

for the account management of the users of the EHR

Cloud. When a user is registered their information is

stored in the Identity manager. Keycloak (Keycloak,

2020) is used as the Identity manager, an Identity

and Access management service which is integrated

with the MinIO object store. The Identity Manager is

also responsible for providing a temporary account to

an HCP during an emergency, in order to access the

user’s EHR. The policy under which the temporary

account is created allows the HCP to only download

the objects that are stored to the EHR Cloud, but can-

not modify them.

The HCP has the ability to add new information to

the EHR of the user, but these additions are kept in a

separate bucket in the EHR Cloud. The content added

by the HCP is only merged with the user’s EHR after

the user reviews and accepts it.

HEALTHINF 2021 - 14th International Conference on Health Informatics

438

Figure 2: EHR Cloud architecture.

4.3 The HCP Certiﬁcation Checker

The HCP Certiﬁcation checker mechanism regards

a service that is only triggered in emergency situa-

tions and speciﬁcally when an HCP requests access

to the EHR Cloud. In order to ensure that the individ-

uals that are certiﬁed as HCPs (doctors, nurses, etc.),

this component requests from the HCP the provision

of the credentials that can be used to verify their oc-

cupation from a trusted authentication authority. As

soon as these credentials are received, the HCP certi-

ﬁcation checker veriﬁes that the individual is indeed

an HCP and the access to the EHR Cloud is granted.

4.4 The Auditing Mechanism

Finally, the Access Auditing component is a Mon-

goDB (MongoDB Inc., 2020) database that keeps logs

of all actions performed in an EHR by the user them-

selves or an HCP during an emergency. In more de-

tail, what is stored in the Auditing component has

to do with the registration of the user, the time that

content is uploaded to the EHR Cloud, and changes

that are made to the EHR by the user. In addition,

logs concerning the HCPs that access the EHR Cloud

are also kept, including the list of the EHR data ob-

jects they download, and the list of the healthcare data

created during an emergency that are uploaded to the

EHR Cloud.

4.5 Communication Gateway

For the communication between the users and the

healthcare professionals with the EHR Cloud a gate-

way is implemented, using Flask (The Pallet Projects,

2020). This gateway implements the functionalities

that a user (either an individual or an HCP during an

emergency) may perform on the EHR Cloud.

These functionalities can be split into two main

categories: the functionalities performed by a user

of the EHR Cloud and the functionalities performed

by an HCP during or after an emergency. Regarding

the user, using this gateway they may register, login

to and deactivate their account from the EHR Cloud.

In addition they may use it to upload, download and

update their EHR to the EHR Cloud. Regarding the

HCP, they may request access to an EHR stored in the

EHR Cloud, as well as upload new content once the

emergency is over.

5 EVALUATION OUTCOMES

This section describes the preliminary tests executed

while deploying the above-mentioned architecture.

Healthcare Provision in the Cloud: An EHR Object Store-based Cloud Used for Emergency

439

Figure 3: Average Throughput (download/upload) of the EHR Cloud.

Figure 4: Average latency (in ms) to perform read / write operations to the EHR Cloud.

5.1 Evaluation Environment

In order to evaluate the proposed EHR Cloud, a setup

that consists of 3 Virtual Machines (VM) is created.

Each VM is comprised of 4 vCPU cores, 32GB of

memory and 1TB of storage and static IPs. All VMs

run CentOS 7 and have Docker installed. In each

VM a containerized application is deployed. More

precisely, on the ﬁrst VM a MinIO container along

with a containerized Flask service that handles all in-

coming requests is deployed. In the second VM, runs

the KeyCloak container along with the containerized

HCP certiﬁcation checker service, while in the third

VM runs the MongoDB Auditing service.

5.2 Results

This section describes the results that were derived

after completing the experimental evaluation over the

deployed EHR Cloud.

For this evaluation the main testing point was the

performance of the deployed system with respect to

the latency in order to perform Read / Write opera-

tions taking under consideration also the ﬁle size. It

is particularly important to be able to download EHR

content fast, especially large ﬁles (i.e. medical im-

ages), from the Cloud especially when in emergency,

since in such cases even a few seconds can make a

difference.

For the evaluation of the deployed EHR Cloud

several use cases where designed each one with dif-

ferent numbers of simultaneous users and ﬁle sizes.

More precisely, we tested the behavior of the EHR

Cloud for 1, 5, 10, and 50 simultaneous users and

simulated the behavior for 100 and 1000 simultane-

ous users that perform either Read or Write operations

in order to measure how the number of the users im-

pact on the performance of the EHR Cloud. The size

of the ﬁles that were used were either 25, 50, 500 or

1000 MB.

Figures 3 and 4 present the results after running

experiments on the deployed EHR Cloud. In these

experiments the upload and download of EHR was

simulated with ﬁles of different sizes in order to ob-

serve the behaviour of the deployed system. As seen

in Figure 4 the latency for the execution of Read /

Write operation to the EHR Cloud is irrelevant to the

number of simultaneous users, when this is number is

HEALTHINF 2021 - 14th International Conference on Health Informatics

440

Table 1: Average Throughput (MBps) (Up/Down) with respect to simultaneous users and encrypted EHR size (MB).

EHR Size (MB)

25 50 500 1000 25 50 500 100

Throughput Down (MBps) Throughput Up (MBps)

Users

1 1.558 1.470 1.470 1.919 1.190 1.190 1.234 1.104

5 1.543 1.408 1.466 1.937 1.162 1.196 1.213 0.995

10 1.548 1.399 1.424 1.921 1.126 1.254 1.264 1.002

50 1.515 1.370 1.445 1.601 1.146 1.212 1.166 1.322

100 1.462 1.355 1.470 1.745 1.116 1.244 1.231 1.301

1000 1.582 1.336 1.466 1.773 1.100 1.257 1.205 1.092

low. This number of the other side increases when the

number of simultaneous users increases. But given

the fact that the current infrastructure did not have the

ability to scale the system, the results show no issues

regarding the functionality of the Cloud.

Figure 3 on the other side depicts the throughput

when performing the experiments that are described

previously. No signiﬁcant changes on the throughput

can be observed. More detailed information are de-

picted in Table 1.

6 DISCUSSION

The evaluation outcomes which are presented in sec-

tion 5 shows that the deployed EHR Cloud’s perfor-

mance is not affected by the number of the users that

are using it simultaneously or by the different-sized

data that is uploaded. In addition, no service failures

were identiﬁed during the tests that were executed.

These results however, are related solely to the perfor-

mance of the service and not to the security require-

ments that should be met by an EHR storing cloud-

based service.

As described in the previous sections the main

purpose of the proposed EHR Cloud architecture is

to automate and accelerate the EHR exchange dur-

ing emergency situations. Regardless, the privacy

and security risks should not be undermined. For

this reason, in this section the requirements that ev-

ery EHR Cloud should meet, as identiﬁed by (Chen

et al., 2012) are presented, in combination with how

those requirements are addressed in the proposed

EHR Cloud. An important note is that the analysis

of these requirements was done prior to the design of

the proposed EHR Cloud architecture.

• Ownership of Information: In the proposed

EHR Cloud architecture the owner and managing

entity of an EHR is the user (i.e. the patient).

Therefore, any changes performed by HCPs on

the content stored in the EHR Cloud must be ap-

proved by the owner. As mentioned earlier, new

EHR content created by HCPs is uploaded in the

Cloud and stored in a separate bucket and only if

the user approves this content, is it then perma-

nently moved to the main bucket.

• Authenticity and Authentication: Only authen-

ticated HCPs can access the content of the EHR

Cloud. An HCP should be authenticated by a

trusted Authentication authority before requesting

access to the EHR Cloud.

• Non-repudiation: Non-repudiation is achieved

in the proposed EHR Cloud architecture, since

only digitally signed transactions and modiﬁca-

tions on the content stored in the EHR Cloud are

allowed.

• Patient Consent and Authorization: In the pro-

posed EHR Cloud two consents are accepted by

the user. The ﬁrst one regards the use of the

EHR Cloud as a backup service where the user’s

healthcare data is stored in the EHR Cloud. The

second one regards the authorization to the EHR

Cloud provider to authorize authenticated HCPs

to access the user’s EHR when an emergency oc-

curs. If the second consent is not accepted the

user may use the EHR Cloud solely as a Storage

Cloud service. In addition both consents are dig-

itally signed by both parties (i.e. the user and the

EHR Cloud provider).

• Availability: High availability is crucial to Cloud

systems, especially when concerning emergency

situations. The system should be available at all

times, even when power outages, hardware fail-

ures or denial-of-service attacks are made. Dur-

ing the evaluation we put the EHR Cloud under,

no timeouts or failures were identiﬁed, thus lead-

ing to the conclusion that the 99.9% availability

requirement was met. Of course, additional mea-

sures should be taken when in production in order

to meet this requirement.

• Data Integrity and Conﬁdentiality: Conﬁden-

tiality and data integrity is achieved through the

encryption of the healthcare data before its upload

the EHR Cloud.

• Auditing: A component dedicated to recording

Healthcare Provision in the Cloud: An EHR Object Store-based Cloud Used for Emergency

441

any transactions and every access in the content

stored in the EHR Cloud is included in the pro-

posed architecture in order to ensure that auditing

is done properly.

7 CONCLUSIONS

This paper proposed a Cloud-based EHR approach

used for safely storing EHRs in the Cloud, while au-

thorizing healthcare professionals to have access to

this content when an emergency situation occurs. The

proposed architecture automates the procedure of au-

thentication and access to the EHR Cloud, thus accel-

erating the process of healthcare services provision.

In order to evaluate the performance of the proposed

EHR Cloud, experiments were executed and the pre-

liminary results are presented. In addition, the re-

quirements that every EHR Cloud should meet are

presented, along with the way those are addressed in

the proposed architecture.

It is within our future goals to perform addi-

tional evaluation to the proposed EHR Cloud-based

approach. Moreover, we plan on researching on the

integration of the edge cloud to the current architec-

ture in order to reduce latency issues that may arise.

ACKNOWLEDGEMENTS

The research leading to this result has received fund-

ing from the European Union’s Horizon 2020 re-

search and innovation programme under grant agree-

ment No 826106 (InteropEHRate project).

REFERENCES

Cai, Z., Yan, H., Li, P., Huang, Z.-a., and Gao, C. (2017).

Towards secure and ﬂexible ehr sharing in mobile

health cloud under static assumptions. Cluster Com-

puting, 20(3):2415–2422.

Cao, S., Zhang, G., Liu, P., Zhang, X., and Neri, F.

(2019). Cloud-assisted secure ehealth systems for

tamper-prooﬁng ehr via blockchain. Information Sci-

ences, 485:427–440.

Chen, Y.-Y., Lu, J.-C., and Jan, J.-K. (2012). A secure ehr

system based on hybrid clouds. Journal of Medical

Systems, 36(5):3375–3384.

Cohen, J. K. (2018). Report: Healthcare industry leads

in cloud adoption. The healthcare industry is the

furthest along in cloud adoption, compared to the

ﬁnancial services industry and the energy and utilities

industry, according to a West Monroe Partners report.

https://www.beckershospitalreview.com/healthcare-

information-technology/report-healthcare-industry-

leads-in-cloud-adoption.html. Last checked on

Sep 23, 2020.

Davenport, T. and Kalakota, R. (2019). The potential for

artiﬁcial intelligence in healthcare. Future healthcare

journal, 6(2):94.

Factor, M., Meth, K., Naor, D., Rodeh, O., and Satran, J.

(2005). Object storage: The future building block for

storage systems. In 2005 IEEE International Sympo-

sium on Mass Storage Systems and Technology, pages

119–123. IEEE.

Joshi, M., Joshi, K., and Finin, T. (2018). Attribute based

encryption for secure access to cloud based ehr sys-

tems. In 2018 IEEE 11th International Conference on

Cloud Computing (CLOUD), pages 932–935. IEEE.

Keycloak (2020). Keycloak. https://www.keycloak.org/.

Last checked on Sep 15, 2020.

Kiourtis, A., Mavrogiorgou, A., Vidakis, K., and Kyriazis,

D. (2020). Health record index: Secure access of

cloud-stored healthcare data. Studies in Health Tech-

nology and Informatics, 272:221–224.

Manoj, R., Alsadoon, A., Prasad, P. W. C., Costadopoulos,

N., and Ali, S. (2017). Hybrid secure and scalable

electronic health record sharing in hybrid cloud. In

2017 5th IEEE International Conference on Mobile

Cloud Computing, Services, and Engineering (Mo-

bileCloud), pages 185–190.

Martin, N. (2019). How healthcare is using big data and

ai to cure disease. https://www.forbes.com/sites/

nicolemartin1/2019/08/30/how-healthcare-is-using-

big-data-and-ai-to-cure-disease/#17e97ef45cfc. Last

checked on Sep 15, 2020.

MinIO Inc. (2020). High performance, kubernetes na-

tive object storage. https://min.io/. Last checked on

Sep 15, 2020.

MongoDB Inc. (2020). The most popular database for mod-

ern apps — mongodb. https://www.mongodb.com/.

Last checked on Sep 15, 2020.

Seol, K., Kim, Y., Lee, E., Seo, Y., and Baik, D.

(2018). Privacy-preserving attribute-based access

control model for xml-based electronic health record

system. IEEE Access, 6:9114–9128.

Sreekanth, R., Rao, G. V. M., and Nanduri, S. (2015). Big

data electronic health records data management and

analysis on cloud with mongodb: A nosql database.

International Journal of Advanced Engineering and

Global Technology, 3(7):943–949.

The Pallet Projects (2020). Flask | The Pallets Projects.

https://palletsprojects.com/p/ﬂask/. Last checked on

Sep 15, 2020.

Vinati Kamani, A. (2019). 5 ways cloud computing is

impacting healthcare. https://www.healthitoutcomes.

com/doc/ways-cloud-computing-is-impacting-

healthcare-0001. Last checked on Sep 15, 2020.

HEALTHINF 2021 - 14th International Conference on Health Informatics

442