Semantic Overlapping in Translational Bioinformatics Applied to the

Matching between Clinical Trial Eligibility Criteria and Patient

Needs

Radmila Juric

, Eton Williams

and Inhwa Kim

University of South Eastern Norway, Kongsberg, Norway

Samsung SDS Europe Ltd, U.K.

Keywords: Clinical Trials, Eligibility Criteria, OWL/SWR.

Abstract: Software technologies play an important role in defining clinical trials, their eligibility criteria and recruitment

process, in which patient enrol to a trial if they satisfy eligibility criteria. In this research we address the

problem of semantic overlapping between eligibility criteria and patient needs through a software architectural

model which houses a specific computational model based on reasoning upon the overlapping semantics. The

architectural model is deployed using semantic technologies in order to explore the meaning of the

relationships between trials, eligibility criteria and patient needs. The novelty is in the reusability and thus

converting of the existing conceptual models on deriving eligibility criteria, available in literature, into the

proposed OWL model, which can serve any clinical trial and requirements patients may have. This paper is

written by computer scientists interested in manipulating semantics of data through computational models

using modern software technologies. It serves as an invitation to researchers from the biomedical and

translational informatics to debate the future of software support in managing clinical trials.

1 INTRODUCTION

The complex problem of designing Clinical Trials

(CT) and systemizing patient eligibility, using

software technologies, has been in the focus of

research interest for more than a decade (Kopcke and

Prokosch, 2014), (Ross et al., 2010) (Cimino et al.,

2007) (Shankar et al., 2006). The work in this field is

vast and has resulted in numerous solutions, which

address the complexity in defining the purpose of CT

and patents clinical and personal needs.

Unfortunately, in the third decade of the 21

century,

we still do not have a powerful software solution,

which could bring us closer to resolving the problem

and creating an universal environment for

pharmaceuticals, medical professionals and patients,

to address the problem in its entirety. The reasons are

numerous, but we would like to draw the reader’s

attention to the following three facts.

First, creating CT and matching their eligibility

criteria (ET) to patient needs, through software

technologies, is a transdisciplinary work. It would

require a high level of collaboration across many

disciplines. However, if we expect that software

technologies, which constantly offer innovations in

the way we collect and process data, can help in

resolving this problem, then we should put computer

scientists in charge of a new computational model for

one important reason. We should avoid using old-

fashioned software solutions, which proved to have

success in the past and think that they will bring

progress in future and in this particular problem

domain. Computer science, software engineering and

computational modelling are fast moving disciplines,

which require constant engagement if we wish to use

them properly and successfully in any problem

domain. We believe that processing the data

generated form bioscience research needs new

computational models created by computer scientists

and thus it is worthwhile to investigate if this could

help to create an universal solution for matching the

semantic between CT and their ET and patent needs.

Second, the research on creating and running CT,

using software technologies, is dispersed and

scattered. After 10 years, we may say that it is

confined to the existence of numerous repositories,

including ontologies and vocabulary of terms,

generated through natural language processing (NLP)

(Elkin et al., 2016), for the purpose of storing

knowledge and manipulating it, mostly through

Juric, R., Williams, E. and Kim, I.

Semantic Overlapping in Translational Bioinformatics Applied to the Matching between Clinical Trial Eligibility Criteria and Patient Needs.

DOI: 10.5220/0009406503150322

In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 3: BIOINFORMATICS, pages 315-322

ISBN: 978-989-758-398-8; ISSN: 2184-4305

315

queries (Baader t al., 2018). This is far from

expectations we have in software engineering, where

software application manage relevant data and

computations, based on a specifically designed

computational model for this problem domain. None

of them exist in published work.

Third, the vast knowledge generated from

biomedical and pharmacological sciences is

expanding fast and it is almost impossible to bridge

the gap between these fields and clinical practices. In

spite of talking about translational informatics, since

the late 2000s (Payne et al., 2015), (Butte, 2008),

(Tsafnat et al., 2013) and pushing software

technologies into this field to help to bridge the gap

between biomedicine and clinical practices, we have

not even started thinking on how these new advances

in biomedical science may have an impact on CT. We

can not resolve this problem by keeping on creating

new knowledge-bases, new ontologies, new database

or similar repositories, and performing queries upon

them. This will not take us forward. What we need

are new computational models, which would assist in

collecting and managing the semantic of relevant and

shared data across these complex research fields

(Almami et al., 2016), (Juric et al., 2018), (Juric,

2019). New computational models can perform a

miracle in collecting and managing the semantic of

data and their matching in order to answer any

question we may have across this complex, but

semantically rich research field.

In order to understand our contribution, it is

important to note that we do NOT wish to

• propose a new “software” for dealing with the

problem of creating CT and finding eligible

patients,

• address a fraction or a slice of this problem by

going into details on how we would implement it

using software technologies and

• create a new repository of knowledge (called a

knowledge-base in the past), controlled

vocabularies often associated with ontologies, and

add them to the existing pool of sources available

in this problem domain.

What we wish to promote in this paper is a generic

computational model, based on the reasoning upon

the semantic of requirements in CT and patient’s

eligibility criteria. The ultimate goal would be to

exploit the semantic overlapping between the two and

define, through the reasoning process, either a CT, or

ET or patient’s best possible match with the two in

particular circumstances.

Software engineering solutions, which would

support data sharing across disciplines of

pharmaceuticals, biomedical sciences and clinical

practices, including patient clinical data, in order to

create and manage CT, would require a generic

software architectural model (Tarabi and Juric, 2018)

((Juric, 2020). 13,14). It is essential in specifying

sources of shared data and computational models for

identifying the best matching between a CT and

patient needs.

For proving the concept, we illustrate our proposal

by using Semantic Web Technology (SWT) and its

languages OWL/SWRL for defining the reasoning

process in which data is shared from biomedical

research published in the literature.

The paper is organized as follows. In section 2 we

specify why and how the SWT and its languages can

be used in this problem domain. In section 3 we

highlight similar work which influenced this research.

The proposal of the software architectural model is in

section 4 and section 5 illustrates an example of an

ad-hoc creation of an OWL model using existing

knowledge from conceptual modelling of Ct and ET

available in the literature. In section 6 we outline the

deployment of the proposal and in the last section we

debate results of this research and comment on future

steps in the last section.

2 WHY SWT

SWT and its layered cake has widely been used, since

its standardisation in 2004, for interpreting the

meaning of data available on the Internet. In

biomedical science OWL has been used for building

common ontologies and controlled vocabularies

across domains, enriched with reasoning rules in

SWRL for bringing inference and more semantics to

biomedical repositories. Knowledge presentation

with SWRL enabled OWL ontologies is extremely

powerful. It is description logic which allows

definition OWL classes and their constraints, in the

form of object and data properties, which enable the

definition of all concept and relationships between

them. This leads to numerous possibilities of

(i) using and exploiting SWRL enabled OWL

ontologies in many problem domains, and outside

the web (Juric, 2016)

(ii) creating OWL ontologies which are not controlled

vocabularies. They may still represent relevant

knowledge, but they will never become

knowledge-bases

(iii) reasoning upon the content of oWL ontologies

using SWRL rule for either strengthening its

knowledge or adding inference to it.

C2C 2020 - Workshop on COMP2CLINIC: Biomedical Researchers Clinicians Closing The Gap Between Translational Research And

Healthcare Practice

316

If we wished to avoid building knowledge bases and

still use SWRL enabled OWL ontologies for

manipulating the semantic of data we process, then

computational models, which house SWRL enabled

reasoning upon OWL concepts can bring inference

without having complex knowledge systems in the

background or using AI algorithms for creating and

manipulating inference. In this paper we talk about

software engineering applications of the SWT

technology which is OUTSIDE formal ontologies and

knowledge-bases and as such, might be a

promisingfstart for addressing the problem of CT /ET

and paitent needs.

We would like to use SWRL enabled OWL

ontologies in order to define

a) the semantic specific for defining and

manipulating EC for CT,

b) the way of converting the semantic of existing

solutions, which use different

method/technologies for defining EC into an

OWL model and

c) semantic overlapping between CT and their EC

and patient needs.

This semantic overlapping would create a cradle

for reasoning upon OWL concepts, which gives a

semantically rich pool of all possible combinations of

ET and patient needs. The power of reasoning,

secured through semantics overlapping between

relevant OWL concepts, infers either new individuals

or constraints in OWL through SWRL rules.

For readers interested in exploring the ways SWT

helps in the creation and manipulation of semantic

overlapping in biomedicine, we suggest reading our

previous publications (Almami et al., 2017), (Juric et

al., 2018), (Juric, 2019), (Juric 2016).

3 RELATED WORK

The section illustrates how scattered the research on

CT is and how it is impossible to find a thread

between the publications.

(Shankar et al., 2006) propose a knowledge-based

framework, named Epoch, and tailored it to the

Immune Tolerance Network research consortium in

order to cover a spectrum of clinical trials

management activities, by tracking study participants

and biological specimens processed in trial

laboratories. The role of developed ontologies in

their software architectural model is to conceptualize

knowledge in the relevant CT domain. In (Mucke et

al., 2009) a semantic model for representing items in

CT is proposed. Its purpose is to move away from

known database technologies and model the semantic

of the problem domain differently. However, their

semantic model does not feed any software

application and reasoning is not introduced for

decision making relevant for CT. In (Besana et al.,

2010) the SWT is used for CT recruitment and their

ontology contains data from patient heath records in

order to verify eligibility of patients for CT. A

consumer centric tool from (Pate et al., 2015), named

TrailX, which matches patients to CT uses numerous

sources of data, such as patient health records, Google

health and Microsoft Health Vault. However, the

matching of patient information and CT is done using

Columbus Matching technology, which relies on

NLP with the assistance of the Unified Medical

Language System. In (Damen et al., 2013) we can

read about the PASTEL platform which assists in CT

recruitment, by using the semantic generated through

topic maps and in (Dameron, 2013) the authors show

an OWL model which systemizes the ET with

partially known information. The authors of (Lee et

al., 2010) introduce the MindTrial system which

facilitates specific matches between clinical trial

criteria and patient volunteers, using a set of

ontologies and semantic queries. In (Elkin et al.,

2016) we can read how local clinical trials can be

enhanced with ontologies and Internet of Things, with

assistance of technologies such as natural language

processing.

It is obvious from the paragraph above that

ontologies are used for a variety of purposes and no

ontological model, generated in one study/project has

been used in another. Furthermore, the power of logic

reasoning with SWRL, suitable for ontology

matching, has not been exploited at all, and therefore

if there were a need for matching the semantic of CT

to patient eligibilities, some other technologies are

used. Finally, emantic overlapping, which is essential

in logic inference, enabled with SWT, is also not

used. This shows that the SWT has not been

completely utilized in this problem domain, except

for creating or retrieving knowledge-bases.

The inference, secured by the semantic

overlapping between OWL concepts, are not to be

confused with the term “reuse and overlapping” in

biomedical science, as described in (Maulik et al.,

2017). In the SWT world, the semantic overlapping

is a computational mechanism which secures

reasoning and inference as in (Almami et al., 2016),

(Juric et al., 2018), (Juric, 2019), and therefore its use

might be associated to semantic mapping evolution

known in biomedical ontologies, as described in (Dos

Res et al., 2014).

Semantic Overlapping in Translational Bioinformatics Applied to the Matching between Clinical Trial Eligibility Criteria and Patient Needs

317

4 SOFTWARE ARCHITECTURE

If we wish to propose a solution which would address

problems with CT as described in the introduction,

and use the SWT which would infer the matching

between CT/EC and patient needs, Figure 1 shows the

essential software architectural (SA) model.

Software components in Figure 1 are technology

specific, i.e. we can use Java Servlet and Enterprise

Java beans technology in order to create a

computational model from the proposed architecture.

Therefore the applications generated from the SA

in Figure 1 would have a computational model

consisting of two types of computations:

a) typical transactional processing with SQL

databases in the background (left part in Fig. 1)

b) reasoning with OWL concepts through SWRL in

order to perform matching of CL/EC and patients

needs (right part in Fig. 1).

These types of computations are not very

common, but they are feasible and they do have

applications across many problem domains (Juric,

2016). For readers interested in software engineering

aspects of the implementations of applications from

the software architecture in Figure 1, we suggest

sources similar to (Patadia et al., 2011), (Shojanoori,

2013) (Tarabi and Juric, 2018).

SA In Figure 1 is component based and layered,

and allows a synergy between computations with

SQL in Java environments and reasoning with SWRL

in the OWL environment though OWL-API.

Obviously, UI.EC+CT interface (left part of the SA

model) would lead us towards the categorization of

EC and CT, which could be converted into

ontological concepts to secure reasoning with SWRL

for the matching with patient needs. (right part of the

SA model) Semantic Overlapping between CT/EC

and patient needs.

There is one important aspect of the proposed SA.

It will generate a software application suitable for

patients. The CATEGORISATION EC+CT

computations (servlet, left side of Figure 1)) would

collect relevant information from Biomedical

Sources and perform the categorisation of EC for

each CT in order to create an environment for

ontological matching. However, patient requests and

needs have already been categorised through the

PERFORM CTM computations (servlet, right side of

Figure 1) and asserted in the application though

UI.CTM interface (Kataria and Juric, 2014).

However, we could have turned the SA around and

start with categorisation of PATIENTS requests (left

side of Figure 1) and create an environment for

matching with EC for all CT, which have already

been categorised and enter into the PERFORM CTM

computations (right part of Figure 1.) The SA and

computations remain the same, only data and entry

mechanisms with UI change. To summarise, the SA

form Figure 1 creates software application, for

pharmaceutics/ clinicians/ patients, in order to secure

the best possible definition of eligibility criteria and

patient needs.

Figure 1: Software Architecture for enabling ontological

matching (CTM stands for Clinical Trial Matching).

4.1 Semantics of Biomedical Resources

Repository named Biomedical Resources from

Figure 1 is an abstract repository illustrated in Figure

2. There exists a vast biomedical knowledge in

various formats, available through a variety of

software solutions such as

(i) Public databases which exist across Biomedical

Informatics, including potential repositories

where various CT and their EC are advertised.

We should also look at sources such as UMLS

(UMLS) and SemDB (Kilicoglu, 2012) which

describe biomedical semantics and co-relation

between therapeutic drugs and diseases.

(ii) Numerous data sets which are generated from

biomedical experiments, and some of them might

be publically available. They are often associated

with the computational analytics and could be

used with popular leaning and predictive

technologies;

(iii) Biomedical data which available on the web /

social media / dedicated web applications, for the

C2C 2020 - Workshop on COMP2CLINIC: Biomedical Researchers Clinicians Closing The Gap Between Translational Research And

Healthcare Practice

318

purpose of disseminating advances in biomedical

research through either publications or social

media means (blogs, twitter, Facebook).

(i)-(iii) are collated in the red box in Figure 2. It is

important to note that Extraction of Biom. DATA

does not mean pure data retrieval. It should include

both: extraction of relevant data and their semantics

at the same time (Kataria and Juric, 2010), (Saaidi et

al., 2010).

Figure 2: Resources of Biomedical Data.

Figure 2 can also be implemented as a software

application using java technologies (Kataria and

Juric, 2010), but the creation of Biomedical

Resources repository from Figure 1, outlined in

Figure 2, is outside the scope of this work. It would

depend on each particular CT, and would vary

between suitable sources. The options on reusing

biomedical knowledge across a variety of repositories

must be open for future work.

In the study we show an example of using public

databases focusing on CT, but we do not exclude data

collected from publications in which categorisation of

EC for a particular CT is debated.

4.2 OWL Model for CT/ET

Figure 3 shows a potential benefit of using the

semantic from UMLS in order to categorise

biomedical knowledge relevant to CT and convert it

into OWL concepts. Sematic Predications (Ahlers et

al., 20107), (Zhang, 2014), (Machado, 2015) have

already been exploited in biomedical science and

could be an excellent starting point for categorizing

semantic of CT/EC and patient needs.

In Figure 3 we show the pathway from the

predications in the form of triplets:

(subject, predication, object)

where we can define the relationships between

subjects and objects through predications. It is

important to note, that a set of triplets: (s,p,o) could

be directly converted into RDF triplets and

consequently create an ontological model according

to the SWT stack.

Figure 3: From Triplets to OWL Model.

Set {C

,…C

} from Figure 3 are OWL classes,

converted from subjects and objects of semantic

predications. Predicates are converted into OWL

constraints, such as object properties. Ontological

hierarchies, shown at the bottom part of Figure 3 are

the main OWL concepts. Potential object properties

might be of either the “is-a” or “has” format or any

other type of relationship which may have existed

between subject and object in definition of CT. In this

particular example, where we need to define a CT

through its EC, “has” object property is more suitable

for explaining the semantic of a CT through EC

In Figure 3 C

denotes a CT and a set of {O

1,1

,…

1,n

} denote a categorised EC for that C

If we wish to perform semantic matching between

CT/ET and patient needs then ontological structures

Semantic Overlapping in Translational Bioinformatics Applied to the Matching between Clinical Trial Eligibility Criteria and Patient Needs

319

based on triplets (s,p,o) should be available for

describing patient requirements for CT. Therefore

Fig. 4 mirrors Figure 3: P

denotes a patient and a set

of {R

1,1

,… R

1,n

} denote his/her requirements for a CT.

The similarity between Fig. 3 and 4 means that if

we wish to match patient requests with CT/EC then

we should use similar categorisation for both: C

classes are described though {O

1,1

,… O

1,n

} and P

classes though {R

1,1

,… R

1,n

Figure 4: OWL Model for Patient Requirements.

5 REASONING PROCESS

We illustrate the way of identifying semantic triplets

for categorizing the complexity of EC in CL as

described in (Ross et al., 2010). They have a specific

categorization of criteria (EC), affected by complex

relationships between disease diagnosis, clinical

phenotypes, which in turn are refined by their

severity, associated complications or response to a

specific treatment. We use a data sample from (Ross

et al., 2010) in order to illustrate that existing

knowledge (publication) is reusable, i.e. their

categorisation of EC could b entered into our OWL

model as individuals and properties, as long as we can

find semantic triplets (s,p,o).

Figure 5: A selection of triplets for OWL model.

A set of triplets could relate symptoms and

disease, treatment/intervention for the disease,

behaviour of a patient, clinical content related to the

disease, temporal criteria and similar. Predications

are available and range from “caused-by”,

“described-with” and “diagnosed-by”, to “without”,

“at least n times per week”, “at hospital discharge”

and “contains-normal-values”.

Figure 5 shows a selection of triplets, where p1 is

an object/subject representing a patient. Predicates

are defined as object properties between individuals

of classes. Individuals of these classes are

subject/objects defined in these triplets.

If we connect all individuals of these OWL classes

(subject/objects) with appropriate object properties

derived from predicates, we could run SWRL

reasoning upon such classes and create an answer if a

particular patient would be eligible for a CL. The

decision will depend on the reasoning process in

which individual(s) of the PATIENT class will be

eligible for a CT only if the reasoning with SWRL

confirms that the particular patient(s) have the same

object properties (predicates) defined between

him/her (them) and eligibility of a clinical trial

(subjects and objects in triplets).

Figure 6: The Reasoning Process.

The reasoning process from Fig 6 is self

explanatory. Blue arrow specifies definition of object

properties between C

and P

classes, black lines

indicate classes involved in reasoning and red one-

directional arrows show inference: only suitable CT

are moved into the CT_Result class for a patient and

only patients which satisfy the EC would be moved

to Patient_List class. The reasoning process is

programed through the following SWRL rules.

CLINICAL_TRIAL (?, D) and has_EC1(?D, EC1)

and has_ECn (?D,ECn) - > CT_Result (?D).

PATIENT (?, XXX) and has_Req1(?XXX, Req1)

and has_Rwqn (?XXX,Reqn) - > PATIENT_LIST

(?XXX).

where EC

, …. ECn and Req

, …. Req

are predicates

(object properties) identified in triplets from the

above and D/XXX are variable for patient/CT.

The message from the OWL model and reasoning

with SWRL, based on semantic overlapping between

ET for a CT and patient requirements, is that the

semantic overlapping secures an almost instant

(p1, diagnosed, disease1)

(p1,exludes, disease2)

(lab-tests, contains-normal-values, p1)

(disease2, treated-by, antibiotics)

(treatment2, unsuccessful-for, p1)

(visits-repeated, required-for, tretament1)

1, refuses, re

eated visits

)

C2C 2020 - Workshop on COMP2CLINIC: Biomedical Researchers Clinicians Closing The Gap Between Translational Research And

Healthcare Practice

320

answers to the questions we may have. We filtered all

the patients (?D) which satisfy ET for a CT for SWRL

rule 1. SWRL rule 2 answers this question: Which CT

are available for a particular patient? The

computational model remains the same, and SWRL

rule uses the same object properties (predicates). The

only difference would be in the format of the rule:

6 CONCLUSIONS

This paper is written from the computer science and

software engineering perspectives and therefore should

show concepts upon which we can build an application

for finding semantic overlapping between CT and

patient needs. Data sharing and computations upon

biomedical and clinical data will reside within one

dedicated light-weight software application. It should

be, suitable for running in mobile and wireless

environments, where updates and constant changes are

welcome and not seen as obstacles. The application

will NOT build an excessive new knowledge base, but

will reuse existing biomedical knowledge, which has

been growing rapidly on a daily basis and add value to

the way we conduct CT.

This study is one of many attempts of using the

SWT and ontological modelling for the purpose of

creating semantic overlapping for matching of CT and

their EC with patient needs. The OWL model and its

constraints, as introduced in (Juric, 2019) can fit the SA

model from Figure 1 and individuals needed for the

OWL model, which would secure reasoning, could be

taken from any of the available sources which

document information on CT. If we could represent

this knowledge in the form of semantic predications

(triplets), than the implementation of the application,

defined by Figures 1,2 and 3 would be straight forward.

In cases when we do not have structures of triplets, the

categorization of data for CT and EC and patient needs

should be performed with OWL principles in mind.

SWRL enabled OWL ontologies are very powerful and

they can really make a difference in this problem

domain, considering that the computational model and

the SA allow software solutions which could run on

modern environments, including Android. Therefore,

from the computational science and software

engineering perspectives, there should be no obstacles

in commercializing the proposal.

However, this study may send message to

researchers from the biomedical field. Repositories

which may contribute towards Biomedical Resources

in Figure 1 should be available, accessible and shared,

but never integrated or changed due to their role in

creating semantic matching. We should leave their

manipulation to computer scientists who can make this

proposal operational and commercial. This implies

that variations of possible implementations of the SA

from Figure 1 would keep up with advances in

software technologies, as long as we can interpret the

semantic overlapping between CT and ET with patient

needs.

We are currently looking at the possibilities of

using biomedical data sets for running predictive

analytics with learning technologies in order to predict

possible predications between various ranges of

subjects and objects, applicable to clinical trials. The

proposed computational model remains the same.

REFERENCES

Ahlers, M. Fiszman, D. Demner-Fushman, F. M. Lang, T. C.

Rindflesch, 2007. Extracting Semantic Predications from

Medline, Pacific Symposium on Biocomputing 12:209-

220(2007).

Almami, E. Almami, I., Juric, R, 2016. Knowledge

Dissemination In Biomedical Science: Using

Ontological Reasoning For The Analyses Of Activated

Enzymes In Various Diseases, In Proceedings of the 21

SDPS 2016 conference, December 2016

Almami, E. Almami, I., Juric, R, 2016. Knowledge

Dissemination In Biomedical Science: Using

Ontological Reasoning For The Analyses Of Activated

Enzymes In Various Diseases, In Proceedings of the 21

SDPS 2016 conference, December 2016

Baader, F., Stefan Borgwardt, Walter Forkel, 2018. Patient

Selection for Clinical Trials Using Temporalized

Ontology-Mediated Query Answering WWW ’18

Companion, April 23–27, 2018, Lyon, France,

Besana P, Cuggia M, Zekri O, Bourde A, Burgun A. 2010.

Using semantic web technologies for clinical trial

recruitment. The Semantic Web-ISWC 2010: 9th

International Semantic Web Conference, ISWC 2010;

November 7-11, 2010; Shanghai, China. Springer; 2010.

p. 7748.

Butte AJ, 2008. Translational bioinformatics: coming of age.

J Am Med Inform Assoc 15(6):709–714.

Damen, D., Kim Luyckx, Geert Hellebaut, Tim Van Den

Bulcke. Åastel 2013. A Sematic Platform for Assisted

Clinical Trial Patient Recruitment, In Proceedings of the

2013 IEEE Intern. Conference on Healthcare

Informatics, 9-11 September 2013, Philadelphia, PA,

USA.

Dameron, O., Besana, P., Zekri, O., Bourde, A., Burgun, A.,

Cuggia, M. (2013) OWL Model of clinical trial eligibility

criteria compatible with partially-known information,

Journal of Biomedical Semantics, BioMed Cnetral, 2013,

4(1), pp 17.

Dos Reis et al. 2014. Understanding Semantic mapping

evolution by observing changes in Biomedical

Ontologies, inJournal of Biomedical Semantics,

47(2014) pp 71-82.

Semantic Overlapping in Translational Bioinformatics Applied to the Matching between Clinical Trial Eligibility Criteria and Patient Needs

321

Elkin, P.L., SCHLEGEL, D.R. ANAND, E. 2016. Recruiting

Participants to Local Clinical Trials using Ontology and

the IoT, in Studies in Health Technology and Informatics,

January 2016

Juric R. (2019) Semantic matching between Eligible Paitents

and clinical Trials using SWRL enabled OWL

Ontologies, in Proceedings of the 24

–SDPS 2019

Conferece, Taiwan, July.

Juric, R. 2016. Could Semantic Technologies Create a new

Computational Model outside Semantic Web, in

Proceedings of the 21

SDPS Conference, December,

FL, US.

Juric, R., Kim, I. 2017. Software Architectures for Smart

Applications Which Merge Ontological Reasoning With

Big Data Analytics, In Proceedings of the 22

SDPS

2017 conference November 2017

Juric, R., Almami, E. Almami, I. 2018. Semantic

Overlapping of OWL Models in Biomedicine, In

Proceedings of the SDPS 2018 Workshop on

Accountability of AI, December 2018, Bologna, Italy

Kataria, P., Juric, R. 2010. Creating Semantics From User

Inputs Through Ontological Reasoning, in Proceedings

of the 15

International Conference SDPS 2010,

University of Texas in Dallas, US, June 6-11.

Kataria, P., Juric, R. 2014 Transferring Ontological

Individuals In OWL/SWRL Enabled Ontologies, In

Proceedings of 18th International Conference SDPS

2013, Campinas, Sao Paulo, Brazil, October 27-31.

Kilicoglu H, Shin D, Fiszman M, Rosemblat G, Rindflesch

TC. 2012. SemMedDB: a PubMed-scale repository of

biomedical semantic predications. Bioinformatics

2012;28:3158–60.

Köpcke,F., Hans-Ulrich Prokosch, 2014. Employing

Computers for the Recruitment into Clinical Trials: A

Comprehensive Systematic Review, J Med Internet Res.

2014 Jul; 16(7): e161.

Lee Y, Dinakarpandian D, Katakam N, Owens D. (2010)

MindTrial: An Intelligent System for Clinical Trials,

AMIA Annu Symp Proc. 2010;2010:442–6.

http://europepmc.org/abstract/MED/21347017.

Machado, C. M.; Rebholz-Schuhmann, D.; Freitas, A. T.;

Couto, F. M. 2015. The semantic web in translational

medicine: current applications and future directions.

Briefings Bioinf. 2015, 16, 89.

Maulik R. Kamdar, Tudorache, T., Musen, Daniel R., 2017.

A Systematic Analysis of Term Reuse and Term overlap

across Biomedical Ontologies, in Semantic Web 2017;

8(6): 853–871.

Moore, R., Lopes, J., 1999. Paper templates. In

TEMPLATE’06, 1st International Conference on

Template Production. SCITEPRESS.

Mucke, R., Mathias Lobe, Magnus Knuth, Frank Loebe,

2009. Semantic model for Representing Items in Clinical

Trials. Proceedings of the 2009 22nd IEEE International

Symposium On Computer-Based Medical Systems, 2-5

August 2009; Albuquerque, NM, USA.

Ontology Web Language https://www.w3.org/TR/owl-

features/ and Semantic Web Rule Language, available

atT https://www.w3.org/Submission/SWRL/

Patadia, R.; Kataria, P., Juric, R., Kim, I. 2011. Conceptual

Design Of Semantic Software Applications, in

Proceedings of the 16th International Conference on

System Design and Processing Science, SDPS 2011, Jeju

Island, South Korea, June 12-16.

Patel C, Cimino JJ, Dolby J, Fokoue a, Kalyanpur a,

Kershenbaum a, Ma L, Schonberg E, Srinivas K. 2007.

Matching patient records to clinical trials using

ontologies. Proceedings of the 6th International

Semantic Web Conference, 2nd Asian Semantic Web

Conference, ISWC 2007 and ASWC 2007; Nov. 11-15,

Busan, Korea. Springer; 2007. p. 6297

Payne, P., Embi, R.O. Peter, J. (Eds.) (2015) Translational

Informatics: Realizing the Promise of Knowledge-

Driven Healthcare. Springer, London 2015.

Ross J, Tu S, Carini S, Sim I. 2010. Analysis of eligibility

criteria complexity in clinical trials. AMIA Summits

Transl Sci Proc AMIA Summit Transl

Sci. 2010;2010:46–50.

Saaidi, R., Kataria, P., Juric R. 2011. Semantic Management

of Submissions of Applications for Marketing

Authorisation of Medicines”, book chapter, in: Suh, S.C.,

Tanik, M., Gurupur, V. P. (Eds.) Biomedical

Engineering: Healthcare Systems, Technology and

Techniques (,Springer; 1st Edition (20 Aug 2011).

Shankar, R.D. S.B. Martins, M.J. O’Connor, D.B. Parish,

A.K. Das, 2006. Towards Semantic Interoåerability in a

Clinical Trials Management Systems in Proc of ISWC,

LNCS 4273, pp 901-912, 2006, Springer Verlag, Berlin

2006

SWT Semantic Web Technology SWT Road Map, available

at W3C website https://www.w3.org/2001/10/03-sww-

1/slide7-0.html

Shojanoori, R. (2013) “Towards Formalization of Situation-

Specific Computations In Pervasive Computing

Environments”, PhD Thesis, University of Westminster,

London, UK, 2013

Shankar, R.D., S.B. Martins, M.J. O’Connor, D.B. Parish,

A.K. Das, 2006, EPOCH An Ontological Framework to

Support Clinical Trials Management, in ACM

Proceedings of HIKM 06, November, VA, US

Tsafnat G., Lin F., Choong M.K. (2013) Translational

Biomedical Informatics. In: Dubitzky W., Wolkenhauer

O., Cho KH., Yokota H. (eds) Encyclopaedia of Systems

Biology. Springer, New York, NY

UMLS Reference Manual, National library of Medicine (last

update September 2009) available at

https://www.ncbi.nlm.nih.gov/books/NBK9676/pdf/Bo

okshelf_NBK9676.pdf)

Tarabi, M., Juric, R. 2018. Software Architectures For Smart

Applications In The Management Of Chronic Diseases:

A Case Study Of Reversibility Of Diabetes 2, In

Proceedings of the 51

HICSS Conference January 2018

R. Zhang, M. J. Cairelli, M. Fiszman, G. Rosemblat, H.

Kilicoglu, T. C., Rindflesch, S. V. Pakhomova, G. B.

Melton, 2013. Using semantic predications to uncover

drug–drug interactions in clinical data. Journal of

Biomedical Informatics 49 (2014) 134–147

C2C 2020 - Workshop on COMP2CLINIC: Biomedical Researchers Clinicians Closing The Gap Between Translational Research And

Healthcare Practice

322