Towards Speciﬁcation of Medical Processes According to International

Standards and Semantic Interoperability Needs

Tanel S

oerd

, Kristian Kankainen

2 a

, Gunnar Piho

1 b

, Toomas Klementi

1 c

and Peeter Ross

2 d

Department of Software Science, TalTech, Akadeemia Str. 15A, Tallinn, Estonia

Department of Health Technologies, TalTech, Akadeemia Str. 15A, Tallinn, Estonia

Keywords:

Process Meta-Model, Federated Interoperability, Clinical Knowledge Modelling, Electronic Health Record,

Healthcare Workﬂow.

Abstract:

Models of healthcare processes and workﬂows to support the continuity of health care are an essential research

topic in medical informatics. This research topic is driven by the necessity to enable systems (including se-

mantic) interoperability, to see the consistency of clinical data recorded in electronic health records (EHR)

and to understand retrospectively the clinical pathways that led to these data. We propose a process meta-

model and evaluate its usability by modelling the healthcare concepts and models from the ISO 13940 (system

of concepts to support continuity of care) standard. Our meta-model is developed according to the software

design patterns principles, enabling the formal speciﬁcation of knowledge in a machine-readable format and

preserving the history of these speciﬁcations. Our work contributes to the federated interoperability of health-

care information systems (healthcare enterprise applications), utilising executable meta-models that can map

healthcare data at the semantic (medical knowledge) level, even at run-time.

1 INTRODUCTION

Health care is evolving from episodic (symptom-

triggered) health care towards a continuous, preven-

tive healthcare approach (EIC, 2022). There is a need

for continuous and uniﬁed access to health data not

limited to a particular episode of a speciﬁc health case

at a particular health institution (health data silos).

A signiﬁcant effort has been made to store medical

records electronically (EHR). However, the stored in-

formation does not always reveal the context or the

reason behind taking a blood sample, administering a

drug or using a particular treatment.

The current practice of EHR makes it difﬁcult for

healthcare workers to determine clinical data consis-

tency and understand clinical pathways, especially in

secondary use cases (e.g. medical science and clin-

ical research) of EHRs, where data users know nei-

ther the patient nor the context, as physicians do in

EHR primary use cases (diagnosis and clinical care).

While governments and society constantly push hos-

https://orcid.org/0000-0002-0551-927X

https://orcid.org/0000-0003-4488-3389

https://orcid.org/0000-0002-8260-526X

https://orcid.org/0000-0003-1072-7249

pitals to reduce costs and improve efﬁciency, health-

care costs rise yearly. One possible cost-reducing

strategy is wise health data digitalization towards

beneﬁcial secondary usage (PWC, 2009). However,

this task is complicated because information systems

in healthcare institutions have semantically heteroge-

neous data models and approximately 80% of medical

data are unstructured (Negro-Calduch et al., 2021).

According to Chen et al., three architectural ap-

proaches to enterprising interoperability exist: inte-

grated (based on a common format), uniﬁed (based

on a common standard) and federated (no common

format; instead, parties accommodate each other on

the ﬂy) (Chen et al., 2008). We rule out integrated

and uniﬁed approaches because of historical soft-

ware development; most medical institutions use their

in-house-speciﬁc standards or at least in-house ver-

sions of national and international standards. It is

common knowledge that software systems developed

by independent parties tend to be semantically het-

erogeneous even when based on the same standards

(Halevy, 2005; Mocan et al., 2009). This heterogene-

ity is because standards always have room for inter-

pretation. The redesign and redevelopment of these

systems is too risky and costly. Instead, we strive

for the federated approach, where (semantic) interop-

160

Sõerd, T., Kankainen, K., Piho, G., Klementi, T. and Ross, P.

Towards Speciﬁcation of Medical Processes According to International Standards and Semantic Interoperability Needs.

DOI: 10.5220/0011687900003402

In Proceedings of the 11th International Conference on Model-Based Software and Systems Engineering (MODELSWARD 2023), pages 160-167

ISBN: 978-989-758-633-0; ISSN: 2184-4348

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

erability of systems (Ciampi et al., 2013; Tu et al.,

2016) is achieved by adjudicating each party’s seman-

tic inconsistencies continuously. For this, we are de-

veloping ABC4HEDA (Archetype-based Computing

for Health Data), which is a Single Underlying Model

(Meier et al., 2019) and software that enables Elec-

tronic Health Record (EHR) management in a trans-

parent, secure and semantically coherent way. By

archetype, we mean a business archetype that “is a

primordial thing that occurs consistently and univer-

sally in business domains and business software sys-

tems” (Arlow and Neustadt, 2003), not the archetype

term as utilized by the openEHR (Beale, 2002) and

ISO13606 (ISO, 2019) standards.

ABC4HEDA aims to contribute to more interop-

erable and consistent healthcare data. It is developed

based on software engineering best practices and pro-

poses a theoretically ”standard-ignorant” approach

to the federated semantic interoperability of clini-

cal data models. By “standard-ignorant”, we mean

an approach similar to computing, infrastructure and

platform-ignorant (independent) approaches, where

no uniﬁed standard is deﬁned and third parties can

specify the standards, taxonomies or terminologies

they use declaratively. This paper evaluates whether

the process meta-model in ABC4HEDA is applicable

in the healthcare domain by testing whether it can be

used to declaratively specify the healthcare process of

the ISO 13940 (system of concepts to support conti-

nuity of care) standard (ISO, 2015).

Our approach has been explained in (Piho et al.,

2010a; Piho et al., 2010c; Piho et al., 2011; Piho

et al., 2012; Piho et al., 2014), evaluated in real-life

applications (Piho et al., 2010b) and analysed on the

basis of HL7 RIM (reference information model) and

openEHR RM (reference model) (Piho et al., 2015) as

well as LOINC (Raavel et al., 2022) and HL7 FHIR

(Randmaa et al., 2022) interoperability tests.

The rest of this paper is structured as follows. In

Section 2, we explain the methodology we have uti-

lized. Sections 3 and 4 brieﬂy overview the ISO

13940 standard and Process Archetype Pattern, re-

spectively. Section 5 speciﬁes the healthcare pro-

cess from the ISO 13940 standard in the domain-

speciﬁc language based on the Process Archetype Pat-

tern meta-model of ABC4HEDA. We analyse and dis-

cuss the presented results in Section 6 and conclude

the paper in Section 7.

2 METHODOLOGY

Consider a situation where a clinical guideline speci-

ﬁes how and in which order the particular healthcare

activities should be performed. And consider that

medical science is evolving and the guideline is re-

vised. Based on this revision, the healthcare institu-

tion has to change its practices; therefore, it is neces-

sary to also change the information system to support

the new guideline. However, as the historical data are

based on the old guidelines, the new system must now

support both versions.

Instead of convening a team of developers to

implement such changes to the information system,

in the ABC4HEDA approach, changes are speci-

ﬁed declaratively while the application is running.

Changes can be rolled back, updating the source code

is not required and the application does not need to

be restarted for healthcare providers to proceed with

work and utilise the new guideline.

ABC4HEDA makes use of syntax, semantics

and pragmatics-based three-level modelling (Bjørner,

2006). All data and knowledge are stored in an

archetype and archetype pattern(A&AP)-based model

(syntax – what we write) (Piho, 2011). These

archetypes and archetype patterns have been designed

(Piho et al., 2010c) according to the Zachman Frame-

work (Zachman, 1987) by asking (and answering)

who (involved actors and their roles), what (products

and services), when (date and time), where (location),

why (rules) and how (processes) questions. This paper

evaluates the business process (how) meta-model.

The A&AP model forms a domain-speciﬁc lan-

guage (DSL) (Fowler, 2010) for specifying (not pro-

gramming) all existing and future developed medical

knowledge (taxonomies, ontologies, standards, pro-

tocols, etc.) declaratively (even at run-time) in an un-

ambiguous and machine-readable format (semantics

– what we mean). This DSL is to be applied in a par-

ticular healthcare information system for specifying

(pragmatics – how we use) the healthcare terms and

concepts in use in that particular healthcare organisa-

tion.

The two principal patterns in use – the item de-

scription pattern (Coad, 1992) and the evolving sys-

tems pattern (Oei et al., 1994) – make our meta-model

dynamically evolvable and semantically interopera-

ble despite model changes. In line with the ﬁrst pat-

tern, every concept in the ABC4HEDA meta-model

has a type that is speciﬁable during run-time (Fig-

ure 1, right). And in accordance with the second

pattern, meta-data is stored for the transparency and

history preservation purposes of the items (Figure 1,

left). So, in case of a mistake or an error during a

change in requirements, it is possible to trace and roll

back the system. Consequently, ABC4HEDA sup-

ports second-order evolution (Oei et al., 1994) in such

a way that changes in semantics (meaning of data,

Towards Speciﬁcation of Medical Processes According to International Standards and Semantic Interoperability Needs

161

knowledge about data), which would generally re-

quire a change in meta-models (e.g. database layout,

classes in software), are converted to ﬁrst-order evo-

lution of data because both the data and their seman-

tics (meaning, knowledge) are presented in the same

manner.

Figure 1: Evolving systems (left) and item description.

3 ISO 13940:2015

The ISO 13940:2015 standard “deﬁnes a system of

concepts for different aspects of the provision of

health care” (ISO, 2015, p. 1). It states that the

core business of health care is the interaction be-

tween the patient and healthcare actors. The stan-

dard uses UML (Uniﬁed Modelling Language) dia-

grams to deﬁne around 150 concepts and their in-

terrelations across eight domains: healthcare actors,

healthcare matters, activities, processes, healthcare

planning, time, responsibilities, and information man-

agement. The standard does not have any regulatory

impact on care delivery, but rather gives a conceptual

basis for describing both the content and the context

of healthcare services. We use the standard as a com-

mon healthcare domain model, which, according to

the software triptych principle, “from domain model

via requirements to the software” (Bjørner, 2006) is

wise to use whenever the requirements for informa-

tion systems are analysed. A review (Kankainen,

2021) analysing how the ISO 13940 concepts have

been understood and applied in literature found two

relevant use cases of the standard. The ﬁrst use case

(Oniki et al., 2014) collected information contexts

(discharge diagnosis, cause of death, complication of

surgery, problem list, etc.) from the standard and used

them in data transformation; the same assertive in-

formation is displayed differently in individual con-

texts. The work discusses the importance of having

an ‘implementation layer’, which means that the logi-

cal models are not compiled directly into the artefacts

used in an implemented system; instead, ‘implemen-

tation models’ are used as a buffer between the logi-

cal models and system implementation. Such a tech-

nique allows context-dependent transformations of

the assertive information and implementation-speciﬁc

transformations for performance optimisation. The

second use case (Garc

ıa-de Le

on-Chocano et al.,

2015) exploited the fact that the contexts enumerated

in the ISO 13940 standard hold universally across cul-

tural, national and organisational borders. Contextu-

alising their data quality metrics with the ISO 13940

standard enabled them to replicate their methodology

at a cross-organisational level.

4 BUSINESS PROCESS

ARCHETYPE PATTERN

A large part of the ABC4HEDA model is based

on business archetypes (Arlow and Neustadt, 2003),

which allow us to respond to all the questions in

the Zachman framework. While Arlow & Neustadt

do not deﬁne an archetype pattern for business pro-

cesses explicitly, they describe how to manage re-

lationships between parties through a customer rela-

tionship management archetype. With some amend-

ments to their pattern, we have created the business

process archetype pattern (Figure 2) by generalising

their party relationship archetype (Piho, 2011).

Figure 2: Business process archetype pattern.

In a similar fashion to a ﬁlm being a series of static

pictures, the process archetype pattern speciﬁes the

processes’ dynamics through an ordered set of reports

and feedback from the process. More reports and/or

feedback give a better overview of the entire process

as a dynamic phenomenon. The same pattern can

be used for business process planning by replacing

actual reports and/or feedback with expected reports

and/or feedback. By comparing the actual with the

expected, we can monitor the adequacy of the plans.

The central part of our business process archetype

pattern is the task archetype (Figure.3). A task is a

party relationship between two party roles and allows

us to model multiple simultaneous roles, e.g. in terms

of a family member, a person can simultaneously be

a parent and a spouse, or a daughter and a sister.

This modelling approach combined with the afore-

MODELSWARD 2023 - 11th International Conference on Model-Based Software and Systems Engineering

162

Figure 3: Task archetype.

mentioned patterns for item description and systems

evolution makes it possible to deﬁne new role and re-

lationship types at run-time. Additionally, it allows

us to keep information consistent in ABC4HEDA,

which, according to Lagos et al. (Lagos et al., 2018),

is a crucial aspect of information models and systems.

Tasks (e.g. an appointment) are grouped into

threads (e.g. patient’s clinical examinations to deter-

mine a possible diagnosis), and threads are grouped

into processes (e.g. a patient’s stay in hospital). A

task contains one or many actions (e.g. measuring

blood pressure), and each action may have one or

many outcomes (e.g. both systolic and diastolic value

of a blood pressure). All the process archetype el-

ements (process, thread, task, action and outcome)

have declaratively speciﬁable types that can be added

during run-time. For certain activities, business pro-

cesses often require approval that can only be given

by authorised parties. For example, Estonia has ﬁve

licences for providing various medical services under

the Health Care Services Organisation Act. Legal co-

ordination must take place between the parties, which

is made possible using the party signature archetype

(PartySignature). Finally, TaskRouting allows tasks

to be transferred between parties (e.g. a change in

treating physician).

5 SPECIFYING THE ISO 13940

SYSTEM OF CONCEPTS

In this section, we specify the concepts and relation-

ships from the ISO 13940 standard using the domain-

speciﬁc language of the ABC4HEDA meta-model de-

scribed above. Uniﬁed Modelling Language (UML)

object diagrams are used for visualisation. In the ob-

ject diagrams, the ABC4HEDA meta-model terms are

pictured as classes (types), and the ISO 13940 stan-

dard concepts are pictured as objects (type instances).

5.1 Specifying the Care Plan

The core of a patient’s healthcare process in ISO

13940 is the dynamic, personalised care plan that

encompasses the activities, goals and objectives re-

lated to a health issue (Figure 4). The ’care plan’

is modelled as a Task in the ABC4HEDA business

process meta-model (domain-speciﬁc language). In

ISO 13940, multiple ’care plans’ can address multi-

ple ’health threads’, where a health thread is a deﬁned

association between healthcare matters as determined

by one or more healthcare actors. In ABC4HEDA, we

model health threads as Thread. The ’care plan’ and

’health thread’ are organised by adding health threads

to the healthcare process and adding care plans to the

health thread. According to ISO 13940, the care plan

is the result of ’healthcare planning’ and it is ’health-

care activity management’ that changes the statuses of

healthcare activities in the care plan. In ABC4HEDA,

healthcare planning is a Process and a change in the

status of activities is modelled using the ActionSta-

tus. Care plans are applied by ’healthcare actors’ –

either organisations or persons participating in health

care. In ABC4HEDA, healthcare actors are deﬁned

in terms of their roles (PartyRole) or signatures (Par-

tySignature). The healthcare process is set in motion

by a health issue identiﬁed by one or more health-

care actors. An example of a healthcare process is

a patient’s stay at a hospital until discharged. Dur-

ing the stay, the patient can undergo many ’health

threads’ (Thread). Each Thread is related to a spe-

ciﬁc problem that will be dealt with in hospital clinics

and centres. A ’referral’ in the ISO 13940 standard is

the movement of a ’healthcare commitment’ between

’healthcare professionals’. The same is accomplished

by TaskRouting in ABC4HEDA.

Activities in a care plan can be arranged to com-

ply with a ’clinical guideline’. The ISO 13940 stan-

dard describes a clinical guideline as a systemati-

cally developed statement designed to assist health-

care actors in making decisions about healthcare ac-

tivity related to speciﬁed health issues. Just like other

tasks (healthcare appointments), clinical guidelines

are modelled by speciﬁc types of activity (Action-

Type). These guidelines make up checklists for the

patient’s care path, similarly to Nan (Nan et al., 2015).

Additionally, a ’care plan’ can be based on a ’core

care plan’, which provides reusable content and struc-

ture for a speciﬁc set of circumstances. We model

core care plans as Threads speciﬁed by ThreadType in

Towards Speciﬁcation of Medical Processes According to International Standards and Semantic Interoperability Needs

163

Figure 4: ISO 13940 care plan in ABC4HEDA domain-speciﬁc language.

the ABC4HEDA process model. A care plan aims to

fulﬁl a ’health objective’, and the activities it carries

out target ’healthcare goals’ that contribute to the ob-

jective. How the goals and objectives can be used for

evaluation is described below in the section on health-

care activities.

5.2 Modelling the Healthcare Process

ISO 13940 deﬁnes the ’healthcare process’ (Figure 5)

as transforming the patient’s health. It has an input

health state and an output health state. How the health

state can be observed and assessed is described in the

section on healthcare activities. The healthcare pro-

cess can be evaluated by ’healthcare process evalua-

tion’, where requirements are systematically assessed

against the healthcare processes. The ABC4HEDA

meta-models evaluate the compliance of the process

using RuleContext.

For each healthcare process, a ’healthcare man-

date’ is required. ISO 13940 deﬁnes this as a directive

based on either ’informed consent’ or ’legal authori-

sation’ and a ’healthcare actor’ accepting ’healthcare

commitment’. In the ABC4HEDA meta-models, par-

ties have legal capacity (PartyCapacity) to perform

speciﬁc activities.

The entire process model yields a holistic health

record, which is a data repository regarding both the

health and the health care of the patient. This health

record consists of entity types, along with entities.

Entity types are the planned treatment activities with

the expected results and entities are (immutable) pro-

tocols of correct Action and the associated Outcomes.

5.3 Modelling the Healthcare Activity

The healthcare activity model is shown in Figure 6.

The management of activities within the care plan

was detailed in the above care plan section and not

repeated here. In the ISO 13940 standard, a ’health-

care activity’ targets a ’healthcare goal’, just like the

care plan aims to meet the overall ’health objective’.

A healthcare activity is assessed by the healthcare

evaluation, which refers to the process of evaluat-

ing various aspects of healthcare operations. In our

ABC4HEDA meta-model, when there is a need to re-

view external evaluations or contracts, the rules can

indicate what is needed, and the context of the rules

(RuleContext) indicates what meets the desired goals

and what does not.

Previously, it was pointed out that different man-

dates are needed for the healthcare activities outlined

in the care plan. In this sense, a healthcare activity is

connected to a healthcare commitment. According to

ISO 13940, this refers to the party’s acceptance of a

healthcare mandate to which the healthcare mandate

is assigned. In order to begin any activity and to meet

that condition, the activities included in the Task re-

quire the approval (PartySignature) of both the origi-

nator and the recipient.

ISO 13940 also outlines ’automated health care’,

which is one type of Action (ActionType) initiated by

a responsible healthcare actor and thereafter delivered

by an automatic medical device. The actor as well

as the medical device both represent ’healthcare re-

sources’, as detailed in the next section. The health-

care activities provided to the patient require at least

one of these resources, and at least two parties are in-

volved in a patient’s health-related Task at any given

time.

5.4 Modelling the Healthcare Resource

Health care and its activities need resources (Figure

7). ’Healthcare personnel’ is modelled as Person and

the ’point of care’, i.e. the location at which direct

healthcare activities are performed, is an Organisa-

tionUnit. One shortcoming in the party archetype pat-

tern is that a Party can also be considered a resource

that is consumed or used during the process. As such,

it is likely that the addition of an abstract device to

meta-models will be necessary in the future in order

MODELSWARD 2023 - 11th International Conference on Model-Based Software and Systems Engineering

164

Figure 5: Model of healthcare process using ISO 13940 terms and ABC4HEDA entities.

Figure 6: Model of healthcare activity using ISO 13940 terms and ABC4HEDA entities.

to describe, for instance, automated medical devices.

Resource management and funding for health care

are shown in Figure 7, and both items needed to de-

scribe them are available in the meta-models. For ex-

ample, a Party is needed to fund healthcare resources,

and said Party may have a role in funding (PartyRo-

leType). In addition, the management of healthcare

resources can be performed by a speciﬁc type of Task

(TaskType).

6 DISCUSSIONS AND

CONCLUSION

In ABC4HEDA, around 120K lines of source code

have been written in C# programming language, and

45% of this code consists of automated unit and ac-

ceptance tests to ensure 100% code coverage. The ar-

chitecture and coding practices are clean and approx-

imately 12% of the code is pure POCO (Plain Old

CLR Object), which forms the computable domain

model for the A&AP. Approximately 3% of the code

implements a repository and unit of work patterns

for the platform-independent data-persistent infras-

tructure. The Entity Framework Core as an object-

relational mapper is in use. However, due to clean

code and a clean architecture approach, the mapping

can easily be replaced by another mapper, if neces-

sary. Approximately 40% of the code uses the latest

ASP.NET Core (also replaceable) and provides infras-

tructure and UI features for management and opera-

tion by allowing medical knowledge speciﬁcations to

be written in a declarative manner, therefore enabling

speciﬁcation of medical standards & other commonly

used or custom-made speciﬁcations.

ABC4HEDA follows the federated approach to

the interoperability of systems without establishing

rules and protocols. This standard’s ignorance is

achieved by separating data and knowledge. Rather

than enforcing standards, ABC4HEDA allows clin-

ical knowledge to be speciﬁed declaratively at run-

time. By preserving the history of both data and

knowledge, the integrated systems may evolve inde-

pendently and safely.

We described how the ABC4HEDA meta-

model is capable of encoding terms from the ISO

13940:2015 (system of concepts to support continu-

ity of care) standard. The terms covered all domains:

healthcare actors, matters, activities, processes, plan-

ning, time, responsibilities, and information manage-

ment.

The necessary work behind the meta-model was

done by selecting the clinical guideline for a stroke

patient’s journey created during a design sprint at

North Estonia Medical Centre and devised by the

master’s students of Tallinn University of Technol-

ogy. Several treatment plans were outlined on the

patient’s journey, and the Estonian Health Insurance

Fund provided us with a more detailed description of

the treatment plans. Throughout the research, the pro-

cess model was validated by experts.

We evaluated the usability of the ABC4HEDA

meta-models and software using its domain-speciﬁc

language to specify the ISO 13940 terminology

declaratively. The ISO 13940 standard takes a

Towards Speciﬁcation of Medical Processes According to International Standards and Semantic Interoperability Needs

165

Figure 7: Model of healthcare resource using ISO 13940 terms and ABC4HEDA entities.

process-oriented approach and postulates the inter-

action between patients and healthcare professionals

as the essential aspect of healthcare processes. The

system requirements based on ISO 13940 were visu-

alised and validated using object diagrams.

Our proposed process meta-model contributes to

the semantic interoperability of federated systems and

the inclusion of more quality healthcare data in pri-

mary (care and diagnoses) and secondary (medical

science and research) use cases. Notably, the con-

textualisation dimension of data quality (Aerts et al.,

2021) is increased, as all clinical data in ABC4HEDA

are modelled as process outcomes, which keeps the

content of data (e.g. blood pressure) linked to its con-

text (e.g. the activity of measuring blood pressure and

the reason for performing the activity).

Our process meta-model describes the dynamics

of business processes as a series of conducted and re-

ported tasks as relationships between parties. Process

planning follows the same pattern and allows us to

evaluate the planned with the achieved.

Future work involves validating the order and

party role archetype patterns. The integration of

ABC4HEDA with more healthcare standards and

clinical research standards must also be proven before

the entire business logic can be validated.

AUTHORS’ CONTRIBUTION

TS wrote the manuscript with support from KK,

GP and TK. GP designed ABC4HEDA. TS and two

other bachelor’s students (Rainer Randmaa and Kaur

Matthias Ravel) under the supervision of KK, TK

and GP implemented ABC4HEDA. GP and PR su-

pervised the project and paper writing.

ACKNOWLEDGEMENTS

This work in the project ‘ICT programme’ was sup-

ported by the European Union through the European

Social Fund.

REFERENCES

Aerts, H., Kalra, D., Saez, C., Ram

ırez-Anguita, J. M.,

Mayer, M.-A., Garcia-Gomez, J. M., Dur

a Hern

andez,

M., Thienpont, G., and Coorevits, P. (2021). Is the

quality of hospital EHR data sufﬁcient to evidence its

ICHOM outcomes performance in heart failure? A pi-

lot evaluation. preprint, Health Informatics.

Arlow, J. and Neustadt, I. (2003). Enterprise patterns and

MDA: building better software with archetype pat-

terns and UML. Addison-Wesley, Object Technology

series, Boston, ...

Beale, T. (2002). Archetypes: Constraint-based domain

models for future-proof information systems. In

OOPSLA 2002 workshop on behavioural semantics,

November 4–8, Washington State Convention and

Trade Center, volume 105, pages 1–69, Seattle, Wash-

ington, USA. Citeseer.

Bjørner, D. (2006). Software engineering 1: Abstraction

and modelling. Springer Science & Business Media,

Berlin Heidelberg.

Chen, D., Doumeingts, G., and Vernadat, F. (2008). Archi-

tectures for enterprise integration and interoperabil-

ity: Past, present and future. Computers in industry,

59(7):647–659.

Ciampi, M., De Pietro, G., Esposito, C., Sicuranza, M., and

Donzelli, P. (2013). A federated interoperability archi-

tecture for health information systems. Int. J. Internet

Protoc. Technol., 7(4):189–202.

Coad, P. (1992). Object-oriented patterns. Communications

of the ACM, 35(9):152–159.

EIC (2022). Eic pathﬁnder challenge: Towards the health-

care continuum: technologies to support a radical shift

from episodic to continuous healthcare. Accessed:

2022-03-24.

Fowler, M. (2010). Domain-speciﬁc languages. Addison-

Wesley Professiona, New Tork.

Garc

ıa-de Le

on-Chocano, R., S

aez, C., Mu

noz-Soler, V.,

Garc

ıa-de Le

on-Gonz

alez, R., and Garc

ıa-G

omez,

J. M. (2015). Construction of quality-assured in-

fant feeding process of care data repositories: deﬁni-

tion and design (part 1). Computers in Biology and

Medicine, 67:95–103.

Halevy, A. (2005). Why your data won’t mix: New tools

and techniques can help ease the pain of reconciling

schemas. Queue, 3(8):50–58.

ISO (2015). 13940:2015 Health informatics - system of

concepts to support. International Organization for

Standardization, Geneva. Switzerland.

MODELSWARD 2023 - 11th International Conference on Model-Based Software and Systems Engineering

166

ISO (2019). ISO 13606-1:2019 Health informatics — Elec-

tronic health record communication — Part 1: Refer-

ence model. International Organization for Standard-

ization, Geneva. Switzerland.

Kankainen, K. J. I. (2021). Usages of the contsys stan-

dard: a position paper. In Bellatreche, L., Chernishev,

G., Corral, A., Ouchani, S., and Vain, J., editors, Ad-

vances in model and data engineering in the digitali-

sation Era, MEDI 2021, 21-23 June 2021, Tallinn, Es-

tonia, volume 1481 of Communications in Computer

and Information Science, pages 314–324, Dordrecht

Heidelberg New York London. Springer Nature.

Lagos, N., Mos, A., and Cornax, M. C. (2018). Towards

semantically-aided domain speciﬁc business process

modeling. Data Technol. Appl., 52:463–481.

Meier, J., Klare, H., Tunjic, C., Atkinson, C., Burger, E.,

Reussner, R. H., and Winter, A. (2019). Single un-

derlying models for projectional, multi-view environ-

ments. In Hammoudi, S., Pires, L. F., and Seli

c, B.,

editors, Model-Driven Engineering and Software De-

velopment: 7th International Conference, MODEL-

SWARD 2019, pages 117–128, Prague, Czech Repub-

lic, February 20–22, 2019, Revised Selected Papers.

Springer Nature.

Mocan, A., Facca, F. M., Loutas, N., Peristeras, V., and

Goudos, S. K. (2009). Solving semantic interoperabil-

ity conﬂicts in cross-border e-government services.

International Journal on Semantic Web and Informa-

tion Systems (IJSWIS), 5(1):1–47.

Nan, S., Van Gorp, P., Korsten, H. H., Kaymak, U.,

Vdovjak, R., Lu, X., and Duan, H. (2015). Dccss:

A meta-model for dynamic clinical checklist sup-

port systems. In 2015 3rd International Conference

on Model-Driven Engineering and Software Develop-

ment (MODELSWARD), ESEO, Angers, Loire Valley

France February 9 - 11, 2015, pages 272–279, Se-

tubal. Portugal. SCITEPRESS - Science and Technol-

ogy Publications.

Negro-Calduch, E., Azzopardi-Muscat, N., Krishnamurthy,

R. S., and Novillo-Ortiz, D. (2021). Technological

progress in electronic health record system optimiza-

tion: Systematic review of systematic literature re-

views. International journal of medical informatics,

152:104507.

Oei, J. H., Proper, H. A., and Falkenberg, E. D. (1994).

Evolving information systems: meeting the ever-

changing environment. Information Systems Journal,

4(3):213–233.

Oniki, T. A., Coyle, J. F., Parker, C. G., and Huff, S. M.

(2014). Lessons learned in detailed clinical mod-

eling at intermountain healthcare. Journal of the

American Medical Informatics Association: JAMIA,

21(6):1076–1081.

Piho, G. (2011). Archetypes based techniques for develop-

ment of domains, requirements and software: towards

LIMS software factory. PhD thesis, Tallinn University

of Technology, Tallinn.

Piho, G., Roost, M., Perkins, D., and Tepandi, J. (2010a).

Towards archetypes-based software development. In

Sobh, T. and Elleithy, K., editors, Innovations in

Computing Sciences and Software Engineering, pages

561–566, Dordrecht. Springer.

Piho, G., Tepandi, J., Parman, M., and Perkins, D. (2010b).

From archetypes-based domain model of clinical lab-

oratory to lims software. In The 33rd International

Convention MIPRO, pages 1179–1184, New York.

IEEE.

Piho, G., Tepandi, J., and Roost, M. (2010c). Domain analy-

sis with archetype patterns based zachman framework

for enterprise architecture. In 2010 International Sym-

posium on Information Technology, volume 3, pages

1351–1356, New York. IEEE.

Piho, G., Tepandi, J., and Roost, M. (2011). Evaluation of

the archetypes based development. In Databases and

Information Systems VI, pages 283–295. IOS Press,

Amsterdam.

Piho, G., Tepandi, J., and Roost, M. (2012). Archetypes

based techniques for modelling of business domains,

requirements and software. In Information Modelling

and Knowledge Bases XXIII, pages 219–238. IOS

Press, Amsterdam.

Piho, G., Tepandi, J., Thompson, D., Tammer, T., Parman,

M., and Puusep, V. (2014). Archetypes based meta-

modeling towards evolutionary, dependable and inter-

operable healthcare information systems. Procedia

Computer Science, 37:457–464.

Piho, G., Tepandi, J., Thompson, D., Woerner, A., and Par-

man, M. (2015). Business archetypes and archetype

patterns from the hl7 rim and openehr rm perspectives:

Towards interoperability and evolution of healthcare

models and software systems. Procedia Computer

Science, 63:553–560.

PWC (2009). Transforming healthcare through secondary

use of health data. PriceWaterhouseCoopers.

Raavel, K. M., Kankainen (supervisor), K., and Piho (super-

visor), G. (2022). Introduction of loinc terminology to

archetype patterns based abc4heda base model. BSc

thesis, in Estonian.

Randmaa, R., Bossenko, I., Klementi, T., Piho, G., and

Ross, P. (2022). Evaluating business meta-models

for semantic interoperability with fhir resources. In

HEDA-2022: The International Health Data Work-

shop, June 19-24, 2022, Bergen. CEURAT, Norway.

Tu, Z., Zacharewicz, G., and Chen, D. (2016). A federated

approach to develop enterprise interoperability. Jour-

nal of Intelligent Manufacturing, 27(1):11–31.

Zachman, J. A. (1987). A framework for information sys-

tems architecture. IBM systems journal, 26(3):276–

292.

Towards Speciﬁcation of Medical Processes According to International Standards and Semantic Interoperability Needs

167