A Flexible Semantic Integration Framework for Fully-integrated EHR

based on FHIR Standard

Ahmed Dridi

1 a

, Salma Sassi

, Richard Chbeir

and Sami Faiz

VPNC Lab., FSJEGJ, University of Jendouba, Jendouba, Tunisia

Universite Pau et Pays de l’Adour, E2S/UPPA, LIUPPA, EA3000, Anglet, 64600, France

ISAMM, University of Manouba, La Manouba, Tunisia

Keywords:

Electronic Health Records, Health Data Integration, Semantic Interoperability, FHIR, Web Semantic, IoT

Data.

Abstract:

Despite the huge efforts focused on EHR development and massive many years of widespread availability

of this latter, health care providers and organizations are still looking for innovative solutions to bring in all

IoT data, and unstructured data into electronic health record (EHR) systems. There is a growing need to

semantically integrate health-related data from different sources to support decision-making and improve the

quality of care services provided. In this paper, we propose a ﬂexible semantic integration framework for IoT,

unstructured and structured healthcare data in EHR systems called SF4FI-EHR. It is built on a novel approach

that applying semantic web technologies and the HL7 FHIR standard to handle integration challenges. Our

experiment results from the proof-of-concept study show that the use of such approach does enhance healthcare

data integration as well as overcome obstacles that prevent the optimal exploitation of these data.

1 INTRODUCTION

Over the years, a large amount of health-related data

is produced, captured and usually accumulated con-

tinuously in Electronic Health Records (EHRs) (Men-

achemi and Collum, 2011). EHR is the most widely

used application in the modern healthcare industry,

which has been used with the intent to improve qual-

ity of care and patient outcomes. With the advent of

Internet of Things (IoT) (Ashton et al., 2009), a new

research topic in many academic and industrial disci-

plines, especially in healthcare, has emerged. Wear-

able medical devices connected to the internet, can

help cut costs and improve patient care by collecting

invaluable data that give extra insight on the individ-

uals physical condition. These data collected from

different sources introduce new challenges related to

data integration and processing, due to their structural

characteristics, their heterogeneity, and the lack of se-

mantics. It is worthy to note that the main objec-

tives of the IoT-based healthcare systems cannot be

achieved without consolidate these data using a well-

designed and established methodology so to properly

integrate them with others clinical data into the EHR.

https://orcid.org/0000-0001-6374-1883

Despite the massive effort and investment in

health information systems to make clinical data in-

teroperable and integrable, the widespread use of the

IoT in EHR remains a long-term goal. The need to de-

sign appropriate solutions for seamless and effective

integration of different health data including IoT and

unstructured data into the EHR is more than before.

This paper addresses the problem of health-related

data integration, where data from multiple sources

needs to be aggregated and integrated in a uniﬁed and

rich-semantic representation. Healthcare data integra-

tion is not a trivial task, but its vital to allow clini-

cians to get the whole picture of an individuals health

related-data. Meanwhile, there are some stark chal-

lenges need to be addressed over this task.

The IoT devices continuously generate large

amounts of data that can reﬂect patient’s conditions.

This data is produced in a mix of data formats, mostly,

without a well-deﬁned structure and additional se-

mantic. It has no value in its raw state, so it must be

converted into a uniﬁed representation, and semanti-

cally described to extract high-level knowledge which

can be used to make decisions, as well as to make it

easy to integrate with other data. IoT data integration

in EHR is the ﬁrst challenge we will focus on in this

research.

684

Dridi, A., Sassi, S., Chbeir, R. and Faiz, S.

A Flexible Semantic Integration Framework for Fully-integrated EHR based on FHIR Standard.

DOI: 10.5220/0008981506840691

In Proceedings of the 12th International Conference on Agents and Artiﬁcial Intelligence (ICAART 2020) - Volume 2, pages 684-691

ISBN: 978-989-758-395-7; ISSN: 2184-433X

Another challenge that needs to be addressed is

the unstructured health data integration. Roughly

80% of clinical data is unstructured data and still

largely untapped after it is created. This data is often

recorded as free text, without standard content speci-

ﬁcations, that makes it more difﬁcult to be handled by

EHR systems. Clinician notes, prescriptions, pathol-

ogy and radiology reports, and scanned medical docu-

ments contains meaningful data to deliver high-value

care. Therefore, this useful data needed to be effec-

tively extracted and integrated into the EHR.

In this paper, we are also motivated by enabling

semantic interoperability among structured health

data. These data are standardized by different stan-

dards, encoded with heterogeneous terminologies and

stored in various legacy systems and EHRs. For sub-

sequent seamless integration with other data, these

data need to be retrieved from their original sources

ﬁrst, and then restructured into a common format and

standard terminologies.

Currently, several state of the art solutions, have

tried to solve the issue of interoperability and seman-

tic integration of healthcare data. But, to the best of

our knowledge, still there is no works are able to per-

form such integration efﬁciently, taking into consid-

eration all the challenges aforementioned.

Here, we propose a ﬂexible semantic integration

framework, called SF4FI-EHR, to capture heteroge-

neous health-related data from different sources and

integrate them into an interoperable integrated EHR.

SF4FI-EHR is based on a set of semantic web

technologies and the HL7 FHIR standard (Bender and

Sartipi, 2013) to:

• Process non-standardized health related data

• Create structured and machine-readable informa-

tion and provide means to raw data of various

forms

• Seamlessly integrate data to a consistent descrip-

tion format

The rest of the paper is organized as follows. Sec-

tion 2 introduces main previous work carried-out on

health data integration. In Section 3, we present our

framework. The subsequent sections detail the main

components of the framework architecture. The ex-

perimental results are discussed in Section 4. Sec-

tion 5 describes an example use case to demonstrate

the health data integration process by the proposed

framework. Finally, Section 6 concludes the paper

with suggestions for future work.

2 BACKGROUND AND RELATED

WORKS

Semantic interoperability is still one of the primary

interests in the E-Health community. In this respect,

a lot of efforts have been devoted to achieve semantic

interoperability into EHRs. As a result, several EHR

standards, terminologies, and semantic frameworks

have been proposed in the literature. For instance,

many standards organizations have put countless ef-

forts to introduce standards and speciﬁcations such

as Health Level 7 (HL7) standards, CEN EN 13606,

ISO TC 215, and openEHR (Eichelberg et al., 2005).

These standards, which are currently in continuous

improvement, aim to structure and markup the clin-

ical content for the purpose of exchange (Maharatna

and Bonﬁglio, 2013). A survey and analysis of EHR

standards are presented in (Eichelberg et al., 2005).

Similarly, others attempts (Mori et al., 1998) aims

at providing terminological standards and codes , as

structured lists of terms that give a controlled vocab-

ulary and structured medically pertinent expressions,

covering complex concepts (such as diseases, opera-

tions, treatments and medicines). Despite the signiﬁ-

cant role played by the proposed standards and termi-

nologies in achieving semantic interoperability into

EHR, there is still no common consensus regarding

their adoption.

Consequently, semantic technologies-based inte-

gration frameworks and mediation approaches are re-

quired to deal with multiple health data represen-

tations. To improve the interoperability of EHRs

data, represented with different standards, Sun et

al. (Sun et al., 2015) proposed an approach to in-

tegrate data from heterogeneous resources. RDF

data mapping approach with semantic conversions

between different representations has been applied

to integrate health records from heterogeneous re-

sources and to generate integrated data in different

data formats/semantics to support various clinical re-

search applications. In (Mart

ınez-Costa et al., 2014),

Martnez-Costa et al. described a layered semantic-

driven architecture to improve EHR semantic interop-

erability based on ontologies that formalize the mean-

ing of clinical data. Similarly, a semantic-driven en-

gine called SeDIE, which built on a novel approach

using a statistical method and a Multiple-Criteria

Decision-Making model to overcome the barriers of

integrating unstructured and structured data, is de-

scribed in (Dhayne et al., 2018). A scalable and

standards-based framework for integrating structured

and unstructured EHR data centered on a clinical NLP

pipeline enhanced with an FHIR-based type system,

is described in (Hong et al., 2018). Ozgur Kilic et

A Flexible Semantic Integration Framework for Fully-integrated EHR based on FHIR Standard

685

al. (Kilic and Dogac, 2009) proposed mapping clini-

cal statements between EHRs, that resolve integration

issues by using archetypes, reﬁned message informa-

tion model (R-MIM) derivations, and semantic tools.

The issue of EHR interoperability is also ad-

dressed by numerous international projects includ-

ing; the European project EHR4CR (De Moor et al.,

2015), eMERGE (Rasmussen-Torvik et al., 2014),

SHRINE (Weber et al., 2009), PONTE (Tagaris et al.,

2012), EURECA (Genitsaridi et al., 2015), CDISC

SHARE (El Fadly et al., 2007), and GALEN (Mori

and Consorti, 1998), where authors overcame the in-

teroperability challenges using a platform capable to

bring up semantic interoperability services based on

standard terminologies.

More recently, increasing interests have been fo-

cusing on the integration of the IoT data into EHRs.

As a result, much work has been carried out in or-

der to achieving the interoperability of the sensed and

EHR data. Kumar et al. (Kumar et al., 2016) suc-

cessfully integrated patient glucose data in the EHR

through an Apple HealthKit, which collects data via

Bluetooth, stores them locally on the mobile device

and then sends the data to the EHR system through

the Epic MyChart app running on the Apples mobile

operating system (iOS). In a similar way to Apple

HealthKit, V. Gay and P. Leijdekkers (Gay and Lei-

jdekkers, 2015) developed with third-party partners

a health and ﬁtness Android application called my-

FitnessCompanion, which integrates a wide range of

wearable devices, EHR systems, and other applica-

tions. Vuppalapati et al. (Vuppalapati et al., 2016)

reviewed the role of big data and data analytics in

healthcare. They proposed a sensor integration frame-

work to integrate various sensors data with the elec-

tronic health records. Spark and Apache Kafka are

used for processing a large amount of data in real-

time. To achieve an effective health self-management,

the authors in (Peng and Goswami, 2019) presented

an OWL ontology-based approach aiming to integrate

the health and home environment data from heteroge-

neously built services and devices. Links with exter-

nal ontologies (such as SSN (Compton et al., 2012)

and FOA

), semantic annotation tools and ontology

mapping techniques are used to ensure efﬁcient data

integration. Alamri A. (Alamri, 2018) described a se-

mantic middleware that exploits ontology to support

the semantic integration of IoT and EHR data. Se-

mantic Web technologies and clinical terminologies

are used to dene and normalize the structures and rela-

tionships of the data collected from the IoT healthcare

devices and sensors and data obtained from EHRs.

A clinical decision support system (CDSS) for dia-

http://www.foaf-project.org/

betes self-management is discussed in (El-Sappagh

et al., 2019). The proposed CDSS is based on the

OWL FASTO ontology

, which integrates the seman-

tic capabilities of the SSN (Compton et al., 2012),

BFO (Grenon et al., 2004), FHIR (Bender and Sartipi,

2013), clinical practice guideline (CPG), and medical

terminologies in a uniﬁed manner, in order to provide

an interoperable integration of IoT and EHR data.

3 SF4FI-EHR: SEMANTIC

FRAMEWORK FOR FULLY

INTEGRATED EHR

Ideally, EHRs must capture and integrate data on all

aspects of healthcare over time, with the data being

represented according to relevant controlled vocab-

ularies and standards, enabling consequently a dif-

ferent level of interoperability for the purpose of ex-

change and potential secondary uses.

From this perspective, we present in this pa-

per a ﬂexible semantic framework, called SF4FI-

EHR standing for Semantic Framework for Fully-

Integrated HER, that systematically enables the ag-

gregation of heterogeneous health-related data from

different sources and integrates them as an interoper-

able integrated EHR. SF4FI-EHR intends to integrate

3 categories of health-related data: 1) IoT and med-

ical devices generated data, 2) Unstructured, and 3)

Structured EHR data. The overall architecture of the

proposed framework is depicted in Fig.1.

To deal with the different data categories, a partic-

ular integration process is expected for each one. IoT

raw-data must be preprocessed, modeled, and then se-

mantically annotated by a modular ontology built by

reusing existing knowledge resources. For unstruc-

tured EHR data, an advanced Natural Language Pro-

cessing (NLP) technique is required to identify ac-

tionable insights and generate structured output. For

this purpose, we chose to use MetaMap tool. The

structured EHR data are needed to be retrieved from

their heterogeneous sources and then converted into a

FHIR resources. All collected, preprocessed and stan-

dardized data are converted to FHIR resource formats,

before being mapped to the FHIR OWL Ontology to

construct a comprehensive patient health proﬁle.

As a result, an integrated FHIR-based EHR is cre-

ated and stored within a Relational DataBase for fu-

ture purposes. We chose to use RDB since it is more

popular and more stable, and most of the current EHR

databases are in RDB format.

https://bioportal.bioontology.org/ontologies/FASTO

ICAART 2020 - 12th International Conference on Agents and Artiﬁcial Intelligence

686

Figure 1: The architecture of the SF4FI-EHR framework.

The next subsections describe in detail the key

features of the integration process within the proposed

framework.

3.1 IoT Data Integration

IoT medical devices capable of generating data that

can be integrated into EHRs to make a signiﬁcant

clinical impact and improve the way the patient is

treated by his doctor.

The problem is that IoT raw data streams, sensed

in real time, have no structure and semantics, which

makes their integration with EHR data challenging.

Thus, IoT data must undergo a preprocessing step in

order to: ﬁll in missing values, smooth noisy data and

delete irrelevant values. Afterwards, these data are

converted into a uniﬁed format by applying ISO/IEEE

11073-104zz device speciﬁcations.

In order to use these sensor data effectively, se-

mantic annotation is applied using a modular ontol-

ogy that we called Semantic Medical Sensor Data on-

tology (SMSD). This ontology enables sensor data to

be described with semantic metadata in order to make

them machine-understandable, interoperable, as well

as to facilitate data integration. In order to ensure high

reusability, the SMSD ontology was built by extend-

ing multiple relevant ontologies. Semantic Sensor

Network ontology (SSN) (Compton et al., 2012) has

been extented in SMSD to cope with sensors networks

and data. Furthermore, We extend several concepts

from the FOAF ontology

, VSO ontology (Goldfain

et al., 2011), CONON ontology (Wang et al., 2004),

http://xmlns.com/foaf/spec/

OWL Time ontology

, CoDAMoS ontology (Preuve-

neers et al., 2004) and Units Ontology (Gkoutos et al.,

2012). We use distinguished colors and speciﬁc pre-

ﬁx that indicates the ontology that each concept be-

longs to, in order to differentiate between the reused

concepts and the concepts we have proposed. Fig.2

illustrate these concepts as well as the relationships

between them. The main concepts of this ontology

are:

• SMSD:Patient: concept represents the ”who” in-

formation about the patient. A patient is described

as a person and inherits all the properties of the

class FOAF:Person which is a sub-class of the

of the FOAF:Agent class in the FOAF ontology.

It include all the demographic information neces-

sary to support the administrative, ﬁnancial and

logistic procedures.

• SMSD:Device: concept describes and represents

a type of a manufactured item that is used in the

provision of healthcare. It could be a medical or

non-medical device (Smartphone, Smart Bracelet,

Wearable Glucose Level Monitoring device, etc).

The property contains is used to link between

the SMSD:Device and the SSN:Sensor classes.

SSN:Sensor concept is extended from the SSN

ontology to represent the physical objects that

observe, transforming incoming stimuli into an-

other, often digital, representation(Compton et al.,

2012).

• SMSD:Vital Signs: concept describes

the detected vital signs of patient which

are represented by the four sub-classes

https://www.w3.org/TR/owl-time/

A Flexible Semantic Integration Framework for Fully-integrated EHR based on FHIR Standard

687

Figure 2: The SMSD ontology.

VSO:Blood Pressor, VSO:Respiratory rate,

VSO:Body temperature, VSO:Pulse Rate

which are extended by the VSO ontology that

covers the four consensus human vital signs:

blood pressure, body temperature, respiration

rate, pulse rate.

• SMSD:Context: concept represents the seman-

tics of concepts and their relationships that cap-

tured about a particular context in pervasive com-

puting environments. It provides a simple context

modeling based on four main concepts, which are:

The Activity, the Location, the Time and the En-

vironment:

– SMSD:Activity: represents based on several

sensors readings the patients activity.

– SMSD:Location: represents the physical loca-

tion of the patient.

– SMSD:Time: represents temporal concepts for

describing the temporal properties of the cap-

tured data.

– SMSD:Environment: describes the characteris-

tics of the surrounding environment condition

of the patient.

The SMSD ontology is encoded in OWL 2 (using

Protg 5.5, an open source ontology editor).

3.2 Unstructured EHR Data Integration

It is well known that a lot of relevant data for making

accurate healthcare decisions is available in unstruc-

tured form. This data needs to be converted into a

structured fashion to allow easy analysis.

Advances in artiﬁcial intelligence, such as deep

learning techniques based on Natural Language Pro-

cessing (NLP) (Mellish, 1989), enable easy and

meaningful information extraction from unstructured

data and generate a structured representation.

To make sense of unstructured data, a Named En-

tity Recognition (NER) (Chieu and Ng, 2003) ap-

proach is adopted by the proposed framework. NER

is a sub-ﬁeld of information extraction (IE) that works

by searching for speciﬁc entity terms and by group-

ing them into pre-deﬁned categories. Entity recogni-

tion has been widely developed in the context of the

medical ﬁeld, usually under the name of Medical En-

tity Recognition (MER) (Abacha and Zweigenbaum,

2011) in order to identify medical entities.

MetaMap program (Aronson, 2006) developed by

the National Institute of Health (NIH) is considered a

baseline in MER. SF4FI-EHR uses MetaMap to dis-

cover and map medical text to UMLS terminologies.

MetaMap’s output generated from unstructured data

is formatted in JSON format, and then transmitted to

FHIR Transform Engine (FTE)

to be converted into

FHIR resources. Fig.3 illustrates the described trans-

formation process.

3.3 Structured EHR Data Integration

Structured EHR data are scattered across multiple

sources, designed by different data models, encoded

with heterogeneous terminologies and represented in

http://www.openmapsw.com/products/FTE.htm

ICAART 2020 - 12th International Conference on Agents and Artiﬁcial Intelligence

688

Figure 3: Unstructured EHR data integration process.

various serialization formats. These data need to be

extracted from their original sources ﬁrst, and then

restructured into a common format and standard ter-

minologies, in order to make easy their subsequent

integration with other data.

A RESTful FHIR server based on FHIR Trans-

form Engine (FTE) is embedded into our framework,

to deal with structured EHR data. Since it is REST-

ful, the FHIR server can collect data from distributed

EHR systems via HTTP GET requests. The data get-

ting in response are represented as JSON-based re-

sources thanks to the FTE.

3.4 The Fully Integrated EHR

To consolidate all parts of the collected data into a

single centralized location, and maintain a compre-

hensive health patient proﬁle, a mapping process is

required between the resulted data from the already

described processes and the W3C’s FHIR OWL On-

tology

IoT data, modeled using the SMSD OWL

ontology, are mapped to the FHIR ontol-

ogy using speciﬁc deﬁned mapping rules.

For instance, the SSN:Obersavation and

SMSD:Activity from the SMSD ontology are

mapped to the fhir.ResourceObservation and

fhir.ResourceActivityDefinition, respectively.

For the other pieces of data (resulted from Un-

structured and structured EHR data integration pro-

cesses) which are already standardized based on

http://w3c.github.io/hcls-fhir-rdf/spec/ontology.html

FHIR standard, a direct map between their resources

elements and the FHIR ontology resources is per-

formed. As a result, a FHIR-based fully integrated

EHR that aggregates patient health-related data from

disparate sources, is created. Finally, a conversion

step is running to map FHIR Resources content from

the FHIR OWL ontology to their equivalent in the

FHIR Relational DataBase schema. Data obtained is

stored in FHIRbase, in order to make data available

for future purposes.

4 IMPLEMENTATIONS

As a proof-of-concept implementation, we used: 1)

Protg 5.5 to develop the OWL SMSD ontology for

modeling and semantically annotate IoT raw data, 2)

Snoggle as SWRL-based ontology mapper is used to

assist in the mapping task between the SMSD and the

FHIR ontology, 3) MetaMap tool is used to recog-

nize medical entities from unstructured data and gen-

erate a JSON-based structured output. MetaMap’s re-

sults is transformed into FHIR-based documents for-

mat by using FHIR Transform Engine, 4) RESTful

HAPI-FHIR server based on FHIR Transform Engine

(FTE) is implemented to collect data from distributed

EHR systems, and 5) JSON2OWL mapping is devel-

oped to map JSON -based data to OWL FHIR ontol-

ogy. The collected data are stored in FHIRbase, as a

standard RDB implementation provided by HL7.

5 EVALUATION RESULTS

To validate our work, a clinical data sets relating to

10 patients have been used as input to our framework,

comprising for each patient:

• An standardized electronic health record.

• Non-standardized clinical documents (e.g. bio-

logical analyzes and radiology reports)

• A vital signs data set generated from different

medical devices carried by patient (temperature,

blood pressure, heart rate and respiratory rate)

• Readings from patient smartphone multiple sen-

sors that can indicate the patients’ activities and

his surrounding environment (e.g. GPS, Ac-

celerometer, Ambient Light Sensor, etc.).

The main objective of this evaluation is to exhibit

that the health data integration process can be success-

fully conducted using SF4FI-EHR framework. Each

data set will be processed according to its category.

A Flexible Semantic Integration Framework for Fully-integrated EHR based on FHIR Standard

689

Sensor data are modeled by the modular SMSD on-

tology, unstructured data are processed by MetaMap

and then converted to FHIR resources formats, and

structured EHR data are converted to FHIR resources

formats, before being directly mapped to the FHIR

OWL Ontology. All data collected modeled by the

FHIR ontology are storing into RDB FHIRbase.

A prototype has been deployed that allowed us

to set up a ﬁrst set of tests using the web browser.

SQL queries are executed to demonstrate that the dif-

ferent data resources are successfully integrated to-

gether. The SQL query shown in Fig.4 is executed to

query for a particular patient observation resources.

Figure 4: The query executed.

The query result is presented in Fig.5, in which

we can see that both the blood glucose value from

an unstructured/not standardized document (biology

analysis results) and the vital signs data from medical

devices are both retrieved in the Observation resource

section in the new Integrated EHR.

Figure 5: The result of the query executed.

To make sure that the integration was successfully

carried out, we invited domain experts to assist the

evaluation process and to assess the quality and the

validity of the integrated data. Discussions with ex-

perts led to that the integration data is well done, as

well as the correctness and the consistency of the in-

tegrated data is well ensured.

6 CONCLUSION

In this paper, we described a ﬂexible semantic inte-

gration framework called SF4FI-EHR. Our solution

integrates its data, as well as Unstructured and Struc-

tured EHR data, coming from heterogeneous sources.

The integration of different data formats is achieved

thanks to the application of Semantic Web technolo-

gies. A modular SMSD ontology based on SSN is

proposed to describe its data. MetaMap tool is used

to deal with the unstructured data, in order to extract

meaningful information from these data and generate

a structured output. Historical patient data collected

from distributed EHR systems using a Restful FHIR

server based on FHIR Tranform Engine. All obtained

data are mapped to the OWL FHIR ontology to re-

construct a Fully integrated EHR data. The semantic

interoperability is handled based on the HL7 FHIR

standard.

Future work will focus on evaluating this frame-

work with more real-world use cases for improvement

and validation. Furthermore, We planned to improve

our approach by integrating environmental data such

as Smart Homes and Ambient Assisted Linvig envi-

ronment data.

REFERENCES

Abacha, A. B. and Zweigenbaum, P. (2011). Medical entity

recognition: A comparison of semantic and statistical

methods. In Proceedings of BioNLP 2011 Workshop,

pages 56–64. Association for Computational Linguis-

tics.

Alamri, A. (2018). Ontology middleware for integration

of iot healthcare information systems in ehr systems.

Computers, 7(4):51.

Aronson, A. R. (2006). Metamap: Mapping text to the umls

metathesaurus. Bethesda, MD: NLM, NIH, DHHS,

pages 1–26.

Ashton, K. et al. (2009). That internet of things thing. RFID

journal, 22(7):97–114.

Bender, D. and Sartipi, K. (2013). Hl7 fhir: An agile and

restful approach to healthcare information exchange.

In Proceedings of the 26th IEEE international sympo-

sium on computer-based medical systems, pages 326–

331. IEEE.

Chieu, H. L. and Ng, H. T. (2003). Named entity recogni-

tion with a maximum entropy approach. In Proceed-

ings of the seventh conference on Natural language

learning at HLT-NAACL 2003-Volume 4, pages 160–

163. Association for Computational Linguistics.

Compton, M., Barnaghi, P., Bermudez, L., Garc

ıA-Castro,

R., Corcho, O., Cox, S., Graybeal, J., Hauswirth, M.,

Henson, C., Herzog, A., et al. (2012). The ssn on-

tology of the w3c semantic sensor network incubator

group. Web semantics: science, services and agents

on the World Wide Web, 17:25–32.

De Moor, G., Sundgren, M., Kalra, D., Schmidt, A., Dugas,

M., Claerhout, B., Karakoyun, T., Ohmann, C., Las-

tic, P.-Y., Ammour, N., et al. (2015). Using elec-

tronic health records for clinical research: the case of

the ehr4cr project. Journal of biomedical informatics,

53:162–173.

Dhayne, H., Kilany, R., Haque, R., and Taher, Y. (2018).

Sedie: A semantic-driven engine for integration of

healthcare data. In 2018 IEEE International Con-

ference on Bioinformatics and Biomedicine (BIBM),

pages 617–622. IEEE.

ICAART 2020 - 12th International Conference on Agents and Artiﬁcial Intelligence

690

Eichelberg, M., Aden, T., Riesmeier, J., Dogac, A., and

Laleci, G. B. (2005). A survey and analysis of elec-

tronic healthcare record standards. Acm Computing

Surveys (Csur), 37(4):277–315.

El Fadly, A., Daniel, C., Bousquet, C., Dart, T., Lastic,

P.-Y., and Degoulet, P. (2007). Electronic healthcare

record and clinical research in cardiovascular radiol-

ogy. hl7 cda and cdisc odm interoperability. In AMIA

Annual Symposium Proceedings, volume 2007, page

216. American Medical Informatics Association.

El-Sappagh, S., Ali, F., Hendawi, A., Jang, J.-H., and

Kwak, K.-S. (2019). A mobile health monitoring-and-

treatment system based on integration of the ssn sen-

sor ontology and the hl7 fhir standard. BMC medical

informatics and decision making, 19(1):97.

Gay, V. and Leijdekkers, P. (2015). Bringing health and ﬁt-

ness data together for connected health care: mobile

apps as enablers of interoperability. Journal of medi-

cal Internet research, 17(11):e260.

Genitsaridi, I., Kondylakis, H., Koumakis, L., Marias, K.,

and Tsiknakis, M. (2015). Evaluation of personal

health record systems through the lenses of ec re-

search projects. Computers in biology and medicine,

59:175–185.

Gkoutos, G. V., Schoﬁeld, P. N., and Hoehndorf, R. (2012).

The units ontology: a tool for integrating units of mea-

surement in science. Database, 2012.

Goldfain, A., Smith, B., Arabandi, S., Brochhausen, M.,

and Hogan, W. R. (2011). Vital sign ontology.

Grenon, P., Smith, B., and Goldberg, L. (2004). Biody-

namic ontology: applying bfo in the biomedical do-

main. Studies in health technology and informatics,

pages 20–38.

Hong, N., Wen, A., Shen, F., Sohn, S., Liu, S., Liu, H., and

Jiang, G. (2018). Integrating structured and unstruc-

tured ehr data using an fhir-based type system: A case

study with medication data. AMIA Summits on Trans-

lational Science Proceedings, 2018:74.

Kilic, O. and Dogac, A. (2009). Achieving clinical

statement interoperability using r-mim and archetype-

based semantic transformations. IEEE Transac-

tions on Information Technology in Biomedicine,

13(4):467–477.

Kumar, R. B., Goren, N. D., Stark, D. E., Wall, D. P.,

and Longhurst, C. A. (2016). Automated integra-

tion of continuous glucose monitor data in the elec-

tronic health record using consumer technology. Jour-

nal of the American Medical Informatics Association,

23(3):532–537.

Maharatna, K. and Bonﬁglio, S. (2013). Systems Design

for Remote Healthcare. Springer Science & Business

Media.

Mart

ınez-Costa, C., Kalra, D., and Schulz, S. (2014). Im-

proving ehr semantic interoperability: future vision

and challenges. In MIE, pages 589–593.

Mellish, C. (1989). Natural Language Processing in Pro-

log: an introduction to computational linguistics.

Addison-Wesley.

Menachemi, N. and Collum, T. H. (2011). Beneﬁts and

drawbacks of electronic health record systems. Risk

management and healthcare policy, 4:47.

Mori, A. R. and Consorti, F. (1998). Exploiting the ter-

minological approach from cen/tc251 and galen to

support semantic interoperability of healthcare record

systems. International journal of medical informatics,

48(1-3):111–124.

Mori, A. R., Consorti, F., and Galeazzi, E. (1998). Stan-

dards to support development of terminological sys-

tems for healthcare telematics. Methods of Informa-

tion in Medicine, 37(04/05):551–563.

Peng, C. and Goswami, P. (2019). Meaningful integration

of data from heterogeneous health services and home

environment based on ontology. Sensors, 19(8):1747.

Preuveneers, D., Van den Bergh, J., Wagelaar, D., Georges,

A., Rigole, P., Clerckx, T., Berbers, Y., Coninx, K.,

Jonckers, V., and De Bosschere, K. (2004). Towards

an extensible context ontology for ambient intelli-

gence. In European Symposium on Ambient Intelli-

gence, pages 148–159. Springer.

Rasmussen-Torvik, L. J., Stallings, S. C., Gordon, A. S.,

Almoguera, B., Basford, M. A., Bielinski, S. J., Braut-

bar, A., Brilliant, M., Carrell, D. S., Connolly, J.,

et al. (2014). Design and anticipated outcomes of the

emerge-pgx project: a multicenter pilot for preemp-

tive pharmacogenomics in electronic health record

systems. Clinical Pharmacology & Therapeutics,

96(4):482–489.

Sun, H., Depraetere, K., De Roo, J., Mels, G., De Vloed, B.,

Twagirumukiza, M., and Colaert, D. (2015). Semantic

processing of ehr data for clinical research. Journal of

biomedical informatics, 58:247–259.

Tagaris, A., Chondrogiannis, E., Andronikou, V., Tsatsa-

ronis, G., Mourtzoukos, K., Roumier, J., Matska-

nis, N., Schroeder, M., Massonet, P., Koutsouris, D.,

et al. (2012). Semantic interoperability between clin-

ical research and healthcare: the ponte approach. In

Extended Semantic Web Conference, pages 191–203.

Springer.

Vuppalapati, C., Ilapakurti, A., and Kedari, S. (2016). The

role of big data in creating sense ehr, an integrated

approach to create next generation mobile sensor and

wearable data driven electronic health record (ehr).

In 2016 IEEE Second International Conference on

Big Data Computing Service and Applications (Big-

DataService), pages 293–296. IEEE.

Wang, X., Zhang, D., Gu, T., Pung, H. K., et al. (2004).

Ontology based context modeling and reasoning us-

ing owl. In Percom workshops, volume 18, page 22.

Citeseer.

Weber, G. M., Murphy, S. N., McMurry, A. J., MacFad-

den, D., Nigrin, D. J., Churchill, S., and Kohane, I. S.

(2009). The shared health research information net-

work (shrine): a prototype federated query tool for

clinical data repositories. Journal of the American

Medical Informatics Association, 16(5):624–630.

A Flexible Semantic Integration Framework for Fully-integrated EHR based on FHIR Standard

691