An Holistic Approach to Diagnostic and Therapeutic Care Pathways

Management

Domenico Redavid

1 a

and Stefano Ferilli

2 b

Economics and Finance Department, University of Bari, Largo Abbazia S. Scolastica, Bari, 70124, Italy

Computer Science Department, University of Bari, Via E. Orabona 4, Bari, 70125, Italy

{domenico.redavid1, stefano.ferilli}@uniba.it

Keywords:

Ontologies, Graph Databases, Clinical Pathways.

Abstract:

The European Commission EU4Health program (2021-2027) is launched after the severe health crisis caused

by COVID-19 to support member states in long-term health challenges to build more resilient health systems

aimed at reducing inequalities in access to healthcare. In Italy, the PNRR program has among its goals the

enhancement of the Diagnostic Therapeutic Care Pathways, particularly their complete informatisation to re-

duce the gap currently present at the regional and, in some cases, at the hospital level. This paper describes a

possible AI framework as a starting point for a potential solution to this goal. The proposed solution involves

the use of GraphDB for information persistence and evolved process management methods for the implemen-

tation of Care Pathways.

1 INTRODUCTION

At the European level, there are many community

programs activated to tackle chronic diseases and fu-

ture scenarios of possible non-sustainability of the

system in ensuring adequate levels of care. After

the severe health crisis from COVID-19, the Eu-

ropean Commission launched the EU4Health pro-

gram (2021-2027) (European Union, 2021) to sup-

port member states in long-term health challenges to

build more resilient health systems aimed at reducing

inequalities in access to healthcare. The programme

addresses the following four general objectives:

• Improve and foster health,

• Protect people,

• Access to medicinal products, medical devices

and crisis-relevant products,

• Strengthen health systems.

The last objective, most relevant to our area of com-

petence, contains these speciﬁc objectives:

• Reinforcing health data, digital tools and services,

digital transformation of healthcare,

• Enhancing access to healthcare,

https://orcid.org/0000-0003-2196-7598

https://orcid.org/0000-0003-1118-0601

• Developing and implementing EU health legisla-

tion and evidence-based decision-making,

• Integrated work among national health systems.

In Italy, the National Plan for Chronic Care (PNC)

(Italian Ministry of Health, 2016), issued in 2016 and

being updated in 2024 (Italian Ministry of Health,

2024), stems from the need to harmonise activities

in this ﬁeld at the national level, proposing a doc-

ument, shared with the Regions, which, consistent

with the availability of economic, human and struc-

tural resources, identiﬁes a common strategic design

aimed at promoting interventions based on a uniﬁed

approach, centered on the person and oriented on a

better organisation of services and the full empower-

ment of all care actors. The aim is to contribute to the

improvement of protection for people with chronic

diseases, reducing the burden on the individual, his

or her family and social context, improving quality

of life, making health services more effective and ef-

ﬁcient in terms of prevention and care, and ensur-

ing greater uniformity and equity of access to citi-

zens. Speciﬁcally, the PNC outlines the overall strat-

egy and objectives of the Plan, proposes some lines

of action and highlights the expected results, through

which to improve the management of chronicity un-

der scientiﬁc evidence, the appropriateness of ser-

vices and the sharing of Diagnostic Therapeutic Care

Pathways (Translated from Italian Percorsi Diagnos-

252

Redavid, D. and Ferilli, S.

An Holistic Approach to Diagnostic and Therapeutic Care Pathways Management.

DOI: 10.5220/0013076200003838

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 16th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2024) - Volume 2: KEOD, pages 252-259

ISBN: 978-989-758-716-0; ISSN: 2184-3228

tici Terapeutici Assistenziali (PDTA)). PDTAs, also

known as critical pathways, care pathways, integrated

care pathways, case management plans, clinical path-

ways or care maps, are used to plan and follow a

patient-centered care program systematically. In this

paper, we propose a possible IT framework capable of

handling all aspects of advanced management via AI

of PDTAs.

2 BACKGROUND

In this section, we report the related work and some

speciﬁc prototypes implementation that could be used

to implement the proposed framework.

2.1 Related Works

The importance of clinical pathways has been high-

lighted through the various surveys published in re-

cent years (Du et al., 2020; Emma Aspland and

Harper, 2021), as well as the veriﬁcation methods

adopted from the EU member state (like in (Italian

Ministry of Health, 2023)), but for our purposes we

are interested in approaches involving the use of on-

tologies and their representation (Dissanayake et al.,

2019). In (Bediang et al., 2021) is proposed an ontol-

ogy called Shareable and Reusable Clinical Pathway

Ontology (ShaRE-CP) which integrates four knowl-

edge domains (CP, guidelines, health resources and

context) to make esplicit existing semantic links be-

tween them. It is developed using Semantic Web

languages, in particular OWL2 (W3C OWL Work-

ing Group, 2012), and some speciﬁc ontologies like

OWL-S (W3C OWL Working Group, 2004), to de-

scribe concepts related to process, and Time-OWL

(W3C OWL Working Group, 2022) for manage ac-

tivities related to time in which they are executed.

In (Alahmar et al., 2020), instead, an ontology-based

framework designed to solve the main challenges re-

lated to standardisation, digitisation, and inclusion of

CPs in modern computerised hospitals is described.

The framework is based on an OWL ontology plus

SWRL rules that uses SNOMED CT (Truran et al.,

2010) terminology and includes a coding system that

is speciﬁc to CPs and their artefacts. Much work has

also been done concerning the use of GraphDB ap-

plied to healtcare (Abu-Salih et al., 2023). Speciﬁ-

cally for Health Critical Pathways, ((Naeimaei Aali

et al., 2022) explores the potential of analysing com-

plex clinical pathways using an event log representa-

tion reﬂecting the independent clinical processes. The

event graph, whose creation is realised using Python

and the Neo4J library, is visualised with Graphviz.

This event graph representation allows the user’ anal-

ysis of the relationship between activities of differ-

ent clinical processes, which was not recognisable

in classical process model representation. A work

closer to our proposal is described in (Aldughayﬁq

et al., 2023) where Semantic Web technologies and

GraphDB are combined to implement a knowledge

graph that can be queried with SPARQL to capture

and visualise complex interactions.

In contrast to these approaches, our goal is to build

a system that manages to exploit GraphDB as a repos-

itory of data and relationships between data whose

meaning is speciﬁed through OWL ontologies.

2.2 GraphBRAIN

GraphBRAIN (Ferilli and Redavid, 2020) is a

general-purpose tool that allows designing and col-

laboratively populate knowledge graphs, and pro-

vides advanced solutions for their fruition, consul-

tation and analysis. The GraphBRAIN functionali-

ties are achieved by combining different tasks, tech-

niques and approaches of artiﬁcial intelligence able

to improve knowledge management and (customised)

user experience, including database technology, on-

tologies, data mining, machine learning, automated

reasoning, natural language processing, personalisa-

tion and recommendation, collaborative and social in-

teraction tools, and social network analysis. While

most of these items are investigated and exploited

separately in the state-of-the-art, the relevance of the

GraphBRAIN methodology is in their being really in-

tegrated, and not simply juxtaposed, so that each of

them takes direct or indirectly advantage from all the

others. This allows GraphBRAIN to ﬁnd relevant,

personalised, and non-trivial information, e.g., a so-

cial approach is used to build and integrate ontolo-

gies; user models are used to guide data mining; on-

tologies are used to guide database interaction and

interface generation; data mining is used to ﬁlter a

manageable and relevant portion of a huge graph on

which carrying out automated reasoning, etc. Graph-

BRAIN’s functions can be used online, interactively

by end users or delivered as Web services to other

applications for obtaining selective and personalised

access to the stored knowledge. For the formal repre-

sentation of CPs, a simple taxonomic representation

model is not sufﬁcient, but more expressive models

are needed. For this reason, GraphBRAIN was cho-

sen as it supports multiple knowledge representation

formal languages ranging from OWL (so it is possible

to reuse existing ontologies) to FOL (more ﬂexible for

the variable knowledge structures representation such

as context- and situation-dependent ones).

An Holistic Approach to Diagnostic and Therapeutic Care Pathways Management

253

2.3 WoMan

WoMan (Workﬂow Manager) (Ferilli, 2014) is a

framework for workﬂow learning and management,

based on First-Order Logic representations. Its pro-

cess mining engine, applicable to activity logs com-

ing from actual process executions, can learn mod-

els involving concurrent, repeated, optional and du-

plicate tasks in any combination, also with weighted

elements. Its full incrementality (Ferilli et al., 2013)

avoids the need for having all the examples available

from the beginning, still allowing the learning to start

from scratch. Correct models can be learned using

very few examples (in principle, any set of examples

including at least one representative of each allowed

process). It can also handle noise in a very straight-

forward, intuitive way. Finally, the representation lan-

guage allows the description of not just the ﬂow of

events, but also the context in which the activities take

place, and hence the learning of complex (and human-

readable) pre- and post-conditions for the workﬂow

elements. A relevant issue in Process Management in

general, and in Process Mining in particular, is to as-

sess how well a model can provide hints about what

is going on during the process execution, and what

will happen next. Indeed, given an intermediate sta-

tus of a process execution, knowing how the execution

will proceed might allow the (human or automatic)

supervisor to take suitable actions that facilitate the

next activities. The task of activity prediction may be

stated as follows: given a process model and the cur-

rent (partial) status of a new process execution, guess

which will be the next activity that will take place in

the execution (Ferilli et al., 2017a). WoMan models

can be used for the monitoring and supervision of pro-

cesses and, when applicable, can be translated into

standard representations (like BPMN or Petri nets).

Both controlled and real-world experiments show that

WoMan outperforms existing process mining systems

in accuracy, effectiveness and efﬁciency. It ensures

quick, correct convergence towards the correct model,

using much less training examples than would be re-

quired by statistical techniques, even in the presence

of noise. WoMan is currently being wrapped in a Web

service that can be exploited by external applications

for workﬂows learning, simulation and checking.

3 AN APPLICATION CASE

3.1 Updated PNC

One of the strengths of the updated Italian National

Plan for Chronic Care (PNC) 2024 (Italian Ministry

of Health, 2024) is the deﬁnition of population strati-

ﬁcation tools to be used in health planning. This pro-

vides a speciﬁc framework of chronic disease types

and quantities sorted by geographic areas. Based on

this stratiﬁcation, it is then possible to act in a targeted

manner with digital health techniques that can take

advantage of telemedicine more effectively in accor-

dance with best practices and scientiﬁc evidence re-

specting the reference legislation (i.e., EU, National

and regional). Particularly for Italy, where health

management information systems are regionally es-

tablished and therefore vary, there is a need for an in-

tegrated platform containing information on the char-

acteristics of the assisted population. This is espe-

cially necessary to manage a stratiﬁcation model that

enables the holistic estimation of the different dimen-

sions of care needs with a patient-centred analytical

approach. A strong impulse was given by the tech-

nological evolution and the experience gained dur-

ing the pandemic emergency by COVID-19, which

stimulated the development of telemedicine and digi-

tal health by strengthening especially the tools useful

to improve the quality, effectiveness and efﬁciency of

the services provided to people with chronic diseases.

Based on these factors, the following lines of action

have been identiﬁed:

• Improving the quality, equity, efﬁciency and ap-

propriateness of care through the activation of

care models that combine telemedicine services

with in-person healthcare delivery methods, start-

ing from the health needs of the persons assisted.

• Strengthen health of initiative and promote multi-

disciplinary of interventions through the imple-

mentation of new organisational models and best

practices, also through the development of Artiﬁ-

cial Intelligence tools.

• Strengthening and adapting telemedicine path-

ways to facilitate the taking charge and continu-

ity of care of people with chronic conditions in

the territory, favouring de-hospitalisation and im-

proving the quality of care also through the acti-

vation of innovative organisational models and the

development of digital health.

• Promoting and enhancing the interoperability of

systems, even through corporate interconnection.

• Enhancing training and continuing education

courses in digital health for health professionals.

The expected results concern the implementation of:

• Care models that, in accordance with the indi-

cations of Ministerial Decree 77/2022 (Vinceti,

2023), combine telemedicine services with the

development of regional projects and good

KEOD 2024 - 16th International Conference on Knowledge Engineering and Ontology Development

254

telemedicine practices as a tool to support patient

management.

• New organisational models envisaged by Minis-

terial Decree 77/2022, also through the develop-

ment of digital healthcare, including corporate in-

terconnection and telemedicine.

Through the implementation, treatment with

telemedicine tools of the chronically ill population is

expected to increase.

3.2 PDTA

PDTA stands for Diagnostic Therapeutic Care Path-

way. Pathway means both the patient’s iter, from his

or her ﬁrst contact with the Health System to the ther-

apeutic treatment after diagnosis, and the organisa-

tional iter, i.e. the phases and procedures for taking

charge of the patient carried out by the Health Au-

thority. Diagnostic, therapeutic and care means the

total taking charge of the patient, together with all

those multi-professional and multidisciplinary inter-

ventions that follow. Thus, PDTAs represent speciﬁc

models for a given territory (Corporate for the popu-

lation treated by that speciﬁc regional hospital organ-

isation) that contextualise the Scientiﬁc Guidelines

concerning the organisation of the Health Authority

and the Region, taking into account in the analysis

the available resources and guaranteeing the Essen-

tial Levels of Care (LEA). PDTAs are complex in-

terventions based on the best scientiﬁc evidence and

characterised by the organisation of the care process

for speciﬁc groups of patients, through the coordina-

tion and implementation of standardised consequen-

tial activities by a multidisciplinary team. For several

years, PDTAs have been used to improve the qual-

ity and efﬁciency of care, reduce variability in care

and ensure appropriate care for the greatest number

of patients. These pathways enable practitioners to

act on the appropriateness of therapeutic and care in-

terventions, reorganising and standardising care pro-

cesses and monitoring their impact not only clinically

but also organisationally and economically. PDTAs,

also known as critical care pathways, integrated care

pathways, case management plans, clinical pathways

or care maps, are used to systematically plan and

follow a patient-centred care programme. As re-

ported in (Mincarone et al., 2018), the languages used

to model CP processes are Uniﬁed Modeling Lan-

guage (UML)

and Business Process Model & No-

tation (BPMN)

. These languages are suitable for the

modelling phase but are not optimal for the execution

UML - https://www.omg.org/UML/

BPMN - https://www.omg.org/bpmn/

phase: UML does not provide for execution engines,

whereas BPMN although it has various implementa-

tions, are based on XML-based representations not

designed for the representation of process semantics.

To manage CPs during both phases, a more powerful

representation language is required; to this end, we

propose WoMan as a tool for their management.

3.3 Health Record

In Italy, the Electronic Health Record (Fascicolo Sani-

tario Elettronico - FSE) was established in 2012 and is

deﬁned as the set of health and social-health data and

digital documents generated by clinical events, con-

cerning the patient, referring to services provided by

the National Health Service and also by private health

facilities. With the National Recovery and Resilience

Plan (PNRR), an implementation of the NextGenera-

tionEU program, measures for its enhancement have

been released. The PNRR investment includes:

• The full integration of all health documents and

data types. This will be achieved through the cre-

ation and implementation of a central repository,

interoperability and service platform, the design

of a standardised user interface and the deﬁnition

of the services to be provided by the FSE.

• The integration of documents by regions within

the FSE. This will also be achieved through ﬁ-

nancial support to healthcare providers to upgrade

their technology infrastructure and data compati-

bility, and to regions adopting the FSE platform

through human capital and skills support to im-

plement the changes necessary for its adoption.

Therefore, the objective of the intervention is to fos-

ter the development of a homogeneous FSE through a

technological transformation of information systems

at the national and regional levels to:

• Guarantee a single access point to health services

for citizens and patients.

• Guarantee a single source of information for

health professionals detailing the patient’s medi-

cal history.

• Ensure that Health Authorities, Regions, and the

Ministry of Health have at their disposal tools to

perform data analysis to improve care and preven-

tion.

Concerning the Apulia Region, the health record has

the following ﬁelds: Vaccination certiﬁcate, Medica-

tion supply, Discharge letter, Hospitalisation prescrip-

tion, Pharmaceutical prescription, Specialist prescrip-

tion, Specialist services, Outpatient specialist and

Vaccination card. As can easily be deduced, the type

An Holistic Approach to Diagnostic and Therapeutic Care Pathways Management

255

of information is very varied, including structured and

unstructured text, image text and images of reports

(X-rays, CT scans, etc.), all data that are well suited

to the application of Artiﬁcial Intelligence techniques

ranging from NLP to Image Processing. Persistence

through tools such as Neo4J would therefore be opti-

mal for this purpose, as it can contain all types of data

and represent the relationships between them with a

powerful query engine.

3.4 Issues to Be Tackled

To achieve the objectives set at the ministerial level,

it is necessary to establish at the technical level what

characteristics the representation languages and tech-

niques to be applied must fulﬁl.

1. Combine existing ontologies to represent CP.

There is no accepted standard terminology for CP,

which makes it difﬁcult to represent all aspects of

CP uniformly.

2. Customise a generic CP with a speciﬁc disease.

This will require the use of a hospital CP in use,

the collaboration of experts in the ﬁeld, and the

deﬁnition of speciﬁc modelling steps.

3. Make explicit not formalised medical data and ac-

tivities. CP is often not followed precisely, but it

is important to be able to capture differences that

could lead to the speciﬁcation of valid variants

for a certain territory or the reporting of incorrect

practices.

4. Provide the CP in a format that is understand-

able by medical professionals. Providing a sim-

ple, user-friendly, and logical interface for medi-

cal professionals could be complicated.

Semantic Interoperability can be achieved through the

use of Semantic Web ontologies that enable implicit

information through automatic reasoning. However,

to achieve the expected results, a holistic approach is

required, which is often not achievable by using these

technologies alone. Moreover, the semantics of the

CP must be represented uniformly to that of the data.

In this way, it is possible to carp the possible inﬂu-

ences on the CP as the data changes.

4 PROPOSED APPROACH

In this section, we report how the proposed frame-

work works identifying how prototype implementa-

tion will be used. The approach involves the use of

GraphBRAIN, for the persistence of data and seman-

tics related to both data and CPs, and WoMan, as

the CP management system. In detail, the following

phases can be identiﬁed:

• Import existing ontologies or create new ones.

In this phase, the system administration uploads

the existent upper ontologies describing Health

Record and CP. The imported ontologies must

be harmonised so a semi-automatic alignment or

merge operation involving domain experts will be

required. This operation will allow both the repre-

sentation of semantics in GraphBRAIN’s internal

format, in particular, complex relationships will

be expressed in FOL, and their parts remaining in

the decidable fragment in OWL (see Fig. 1).

Figure 1: Ontology Concepts and Relations Import.

• Import of existing Health Records and PDTA.

The Health Records can be migrated from other

sources to the Neo4J GraphDB used by Graph-

BRAIN. Guidelines for formatting the data in

the structure provided by the GraphBRAIN on-

tology representing the schema of the data in

KEOD 2024 - 16th International Conference on Knowledge Engineering and Ontology Development

256

the graph will be released. Once imported, the

data can be further validated by domain experts

through GraphBRAIN’s front-end interface. Fur-

thermore, existing PDTAs will be formalised in

the WoMan formalism. The transition from the

major languages currently used for CP represen-

tation (BPMN and PetriNet) is already available

among the import features of WoMan itself, for

other representations import is always guaranteed

as FOL has sufﬁcient expressive power to include

the primitives of any logical representation (see

Fig. 2).

Figure 2: Data and PDTA workﬂow Import.

• System Execution. The CP management informa-

tion systems in use provide a web-based front-

end interface from where management and con-

trol features can be performed. These interfaces

may be in an initial phase gradually extended

to use GraphBRAIN and WoMan as a back-end.

Later, these interfaces will be modiﬁed and ex-

tended to provide new functionality. With the

power of these tools, it will be possible to respond

to the issues shown in Sect. 3.4:

1. Combine existing ontologies to represent CP

semantics. GraphBRAIN allows to export of

ontology concepts and relations in several for-

mats, such as OWL, SWRL and FOL

2. Customise a generic CP with a speciﬁc disease.

WoMan allows you to learn new processes from

scratch, observe whether a process is executing

correctly, or learn deviations during the execu-

tion of a process. Therefore, ﬁnding possible

customisations of existing processes and repre-

senting them as variants for a speciﬁc territory

is possible.

3. Make explicit not formalised medical data and

activities. The ability to export portions of

knowledge into OWL allows automatic reason-

ing to be applied to ﬁnd implicit knowledge. In

addition, WoMan’s supervision feature allows

the system to alert physicians and patients if CP

is not being performed correctly.

4. Provide the CP in a format that is understand-

able by medical professionals. The simplicity

of process representation in WoMan allows for

the construction of ad-hoc interfaces that also

report explanations of why an activity is a con-

sequence of previous ones. This will enable

medical professionals to understand the pecu-

liarities of a CP in more detail.

Multistrategy reasoning (Ferilli, 2023) is the key to

being able to ﬁnd appropriate solutions to these prob-

lems. Since GraphBRAIN uses a Labelled Property

Graph as a persistence medium, the following forms

of automatic reasoning can be applied:

• Symbolic/Logic approach: deduction, abduction

(also with probabilistic Reasoning support), ab-

straction, Argumentation for Induction. These

approach can be useful to infer new information

from data (how a patient should be followed up,

what possible CPs are applicable, completeness

of information present on the patient, complete-

ness of medical reports, and so on).

• Statistical/Mathematical approach: Subgraph Ex-

traction with the application of Spreading Activa-

tion or PageRank algorithm, Centrality with dif-

ferent assessment strategies (i.e., Closeness, Be-

tweenness, Harmonic, Katz and Page Rank), Link

Prediction and Clustering. These approaches can

be useful for obtaining statistical information on

the healthcare state at different levels of granular-

ity.

An Holistic Approach to Diagnostic and Therapeutic Care Pathways Management

257

The ability to interoperate with external sources (par-

ticularly GraphBRAIN with OWL ontologies and

WoMan with SWRL rules (Redavid and Ferilli,

2023)) simpliﬁes the acquisition of data and informa-

tion from existing healthcare systems.

Another important issue for telemedicine is the

availability of the back-end support described here.

By using it, it is also possible to capture information

from physician-patient interaction during remote vis-

its, as well as data from computer sensing, which is

critical to the goals to be achieved. The same tech-

nologies proposed in this paper are also able to ad-

dress these needs (Redavid et al., 2022). Different is

the security aspect where, however, good solutions,

especially when we talk about social networks de-

signed for CP consolidation, have been proposed (Pel-

licani et al., 2023).

5 CONCLUSIONS AND FUTURE

WORKS

In this paper, we have outlined a possible framework

that can handle existing or learned PDTA formalised

in the WoMan formalism. Through GraphBRAIN it

will be possible to handle different knowledge repre-

sentations and apply multi-strategic reasoning to im-

prove CP related to Health Records. Combining these

tools we can cover one of the fundamental require-

ments of National Plan for Chronic Care: an holis-

tic approach to managing the different dimensions of

care needs with a patient-centred analytical approach.

In a speciﬁc vision, the implementation of

NextGenerationEU program can be an opportunity to

converge toward a common line of CP management,

in a general vision, the idea that in any territory of the

European community, it is possible to know the treat-

ment path of a citizen of the community leads to a

greater awareness of being part of an evolution repre-

sented by the European Community itself. The holis-

tic approach underlying GraphBRAIN (Ferilli et al.,

2023) as well as the innovative AI approach to process

management (Ferilli et al., 2017b) enables a concrete

response to the problem we have been discussing. The

PNRR AMICA project is a good test case for imple-

menting the proposed approach. In future work, start-

ing from local solutions it will be possible to general-

ize them to seek valid solutions at the European level.

ACKNOWLEDGEMENTS

This work is partially supported by the Ministry of

Health, ’Trajectory 1 ‘Active & Healthy Ageing -

Technologies for Active Ageing and Home Care’ ini-

tiative, funded project: ‘AmICA: Intelligent holistic

care for aCtive Ageing in indoor and outdoor ecosys-

tems’ [Grant number T1-MZ-09].

REFERENCES

Abu-Salih, B., Al-Qurishi, M., Alweshah, M., Al-Smadi,

M., Alfayez, R., and Saadeh, H. (2023). Healthcare

knowledge graph construction: A systematic review

of the state-of-the-art, open issues, and opportunities.

J. Big Data, 10(1):81.

Alahmar, A., Crupi, M. E., and Benlamri, R. (2020). On-

tological framework for standardizing and digitizing

clinical pathways in healthcare information systems.

Computer Methods and Programs in Biomedicine,

196:105559.

Aldughayﬁq, B., Ashfaq, F., Jhanjhi, N. Z., and Humayun,

M. (2023). Capturing semantic relationships in elec-

tronic health records using knowledge graphs: An

implementation using mimic iii dataset and graphdb.

Healthcare, 11(12).

Bediang, G., Falquet, G., and Geissb

uhler, A. (2021).

An ontology-based semantic model for sharing and

reusability of clinical pathways across context (share-

cp). In Otero, P., Scott, P., Martin, S. Z., and Huesing,

E., editors, MEDINFO 2021: One World, One Health

- Global Partnership for Digital Innovation - Pro-

ceedings of the 18th World Congress on Medical and

Health Informatics, Virtual Event, 2-4 October 2021,

volume 290 of Studies in Health Technology and In-

formatics, pages 86–90. IOS Press.

Dissanayake, P. I., Colicchio, T. K., and Cimino, J. J.

(2019). Using clinical reasoning ontologies to make

smarter clinical decision support systems: a system-

atic review and data synthesis. Journal of the Ameri-

can Medical Informatics Association, 27(1):159–174.

Du, G., Huang, L., and Zhou, M. (2020). Variance analysis

and handling of clinical pathway: An overview of the

state of knowledge. IEEE Access, 8:158208–158223.

Emma Aspland, D. G. and Harper, P. (2021). Clinical path-

way modelling: a literature review. Health Systems,

10(1):1–23.

European Union (2021). EU4Health programme 2021-2027

– a vision for a healthier European Union.

Ferilli, S. (2014). Woman: Logic-based workﬂow learn-

ing and management. IEEE Transactions on Systems,

Man, and Cybernetics: Systems, 44(6):744–756.

Ferilli, S. (2023). Gear: A general inference engine for au-

tomated multistrategy reasoning. Electronics, 12(2).

Ferilli, S., Carolis, B. D., and Redavid, D. (2013). Logic-

based incremental process mining in smart environ-

ments. In Ali, M., Bosse, T., Hindriks, K. V., Hoogen-

doorn, M., Jonker, C. M., and Treur, J., editors, Re-

KEOD 2024 - 16th International Conference on Knowledge Engineering and Ontology Development

258

cent Trends in Applied Artiﬁcial Intelligence, 26th

International Conference on Industrial, Engineering

and Other Applications of Applied Intelligent Systems,

IEA/AIE 2013, Amsterdam, The Netherlands, June 17-

21, 2013. Proceedings, volume 7906 of Lecture Notes

in Computer Science, pages 392–401. Springer.

Ferilli, S., Esposito, F., Redavid, D., and Angelastro, S.

(2017a). Extended process models for activity predic-

tion. In Kryszkiewicz, M., Appice, A.,

ezak, D., Ry-

binski, H., Skowron, A., and Ra

s, Z. W., editors, Foun-

dations of Intelligent Systems, pages 368–377, Cham.

Springer International Publishing.

Ferilli, S. and Redavid, D. (2020). The graphbrain sys-

tem for knowledge graph management and advanced

fruition. In Helic, D., Leitner, G., Stettinger, M.,

Felfernig, A., and Ra

s, Z. W., editors, Foundations of

Intelligent Systems, pages 308–317, Cham. Springer

International Publishing.

Ferilli, S., Redavid, D., and Angelastro, S. (2017b). Ac-

tivity prediction in process management using the

woman framework. In Perner, P., editor, Advances in

Data Mining. Applications and Theoretical Aspects -

17th Industrial Conference, ICDM 2017, New York,

NY, USA, July 12-13, 2017, Proceedings, volume

10357 of Lecture Notes in Computer Science, pages

194–208. Springer.

Ferilli, S., Redavid, D., and Pierro, D. D. (2023). Holis-

tic graph-based document representation and manage-

ment for open science. Int. J. Digit. Libr., 24(4):205–

227.

Italian Ministry of Health (2016). Italian National Plan of

Chronicity.

Italian Ministry of Health (2023). Results of NSG-PDTA

indicators for the assessment year 2021.

Italian Ministry of Health (2024). Italian National Plan of

Chronicity Update.

Mincarone, P., Leo, C. G., Trujillo-Mart

ın, M. d. M., Man-

son, J., Guarino, R., Ponzini, G., and Sabina, S.

(2018). Standardized languages and notations for

graphical modelling of patient care processes: a sys-

tematic review. International Journal for Quality in

Health Care, 30(3):169–177.

Naeimaei Aali, M., Mannhardt, F., and Toussaint, P. J.

(2022). Discovering care pathways for multi-morbid

patients using event graphs. In Munoz-Gama, J. and

Lu, X., editors, Process Mining Workshops, pages

352–364, Cham. Springer International Publishing.

Pellicani, A., Pio, G., Redavid, D., and Ceci, M. (2023).

SAIRUS: spatially-aware identiﬁcation of risky users

in social networks. Inf. Fusion, 92:435–449.

Redavid, D. and Ferilli, S. (2023). Semantic web services

ingestion in a process mining framework. Electronics,

12(23).

Redavid, D., Ferilli, S., Loop, L., and Matviichuk, L.

(2022). Emotional states management for an advanced

intelligent tutoring system. In Aveiro, D., Dietz, J.

L. G., and Filipe, J., editors, Proceedings of the 14th

International Joint Conference on Knowledge Discov-

ery, Knowledge Engineering and Knowledge Manage-

ment, IC3K 2022, Volume 2: KEOD, Valletta, Malta,

October 24-26, 2022, pages 253–259. SCITEPRESS.

Truran, D., Saad, P., Zhang, M., and Innes, K. (2010).

SNOMED CT and its place in health information

management practice. Health Information Manage-

ment Journal, 39(2):37.

Vinceti, S. R. (2023). Reorganizing italy’s territorial health-

care: the ministerial decree no. 77/2022 and its com-

parative signiﬁcance. Annali di igiene : medicina pre-

ventiva e di comunita.

W3C OWL Working Group (2004). OWL-S: Semantic

Markup for Web Services - W3C Member Submission

22 November 2004.

W3C OWL Working Group (2012). OWL 2 Web Ontol-

ogy Language Document Overview (Second Edition)

- W3C Recommendation 11 December 2012.

W3C OWL Working Group (2022). Time Ontology in

OWL - W3C Candidate Recommendation Draft 15

November 2022.

An Holistic Approach to Diagnostic and Therapeutic Care Pathways Management

259