Ontology-Based Framework for Personalized Home-Based

Rehabilitation in Cerebral Palsy Care

Rahma Haouas Zahwanie

, Lilia Cheniti-Belcadhi

and Saoussen Layouni

Sousse University, ISITC, PRINCE Research Laboratory, H-Sousse, Tunisia

Physical Medicine and Rehabilitation Department, University of Sousse, Medical Faculty, Sousse, Tunisia

Keywords:

Cerebral Pasly, Rehabilitation, Ontology, Intelligent Framework, Home-Based Exercise Programs.

Abstract:

In the domain of cerebral palsy rehabilitation, advances in machine learning and semantic technologies of-

fer promising solutions to enhance treatment strategies. This paper focuses on developing an ontology-based

framework to support rehabilitation programs for children with cerebral palsy, addressing the need for person-

alized, home-based exercise programs (HEP). These programs aim to improve recovery by enabling patients

to engage in tailored exercises outside clinical settings. However, the effectiveness of HEP depends on ac-

curate monitoring and feedback, as improper execution of exercises can hinder progress. To address this

challenge, we propose an intelligent system framework that integrates ontology-driven knowledge representa-

tion to oversee rehabilitation programs. The system analyzes patient proﬁles and progress data, recommending

a personalized rehabilitation plan consisting of targeted exercises supported by healthcare professionals. The

ontology serves as the backbone of this framework, enabling semantic representation of rehabilitation con-

cepts and facilitating the management and improvement of cerebral palsy treatment pathways. Furthermore,

this approach enhances patient outcomes by providing structured, context-aware rehabilitation plans while

promoting interoperability and knowledge sharing across healthcare systems. By embedding the ontology

within the framework, we enable greater reusability, semantic comprehension, and adaptability to multilin-

gual healthcare environments. This work highlights the critical role of ontologies in advancing rehabilitation

strategies for cerebral palsy and improving access to high-quality, personalized care.

1 INTRODUCTION

Physical therapy is for enhancing the functional abili-

ties of individuals with disabilities or physical impair-

ments. Research emphasizes its role in improving pa-

tient outcomes, showing a strong correlation between

exercise levels and the success of rehabilitation pro-

grams. Rehabilitation therapy is particularly bene-

ﬁcial for patients after hospital discharge following

an acute stroke or cerebral palsy (O’Neill and For-

man, 2020). However, factors such as limited ac-

cess to appropriate healthcare providers, higher costs

that reduce the amount of care provided, and poor

patient attendance limit the beneﬁts of rehabilitation

therapy. Furthermore, even when patients do beneﬁt

from rehabilitation therapy, it is often provided inter-

mittently, with prolonged intervals between sessions,

which limits the effectiveness of the treatment com-

pared with the recommendations of high-quality clin-

ical studies.

Home-based programs are generally recom-

mended as part of rehabilitation programs to offer

greater ﬂexibility, while healthcare professionals cre-

ate individualised rehabilitation plans with exercise

suggestions. Signiﬁcantly, more than 90% of rehabil-

itation programs are carried out at home, demonstrat-

ing the widespread adoption of this technique. (Ko-

matireddy et al., 2014). Studies (Chen et al., 2014)

(Ibeneme et al., 2024) (Ge et al., 2024) have shown

that home exercise can be effective in improving vari-

ous aspects of motor function and quality of life in pa-

tients with cerebral palsy. These studies have shown

that home exercise programs, including tele-assisted

robotic rehabilitation, computer-assisted upper limb

exercises, and individualised interactive training, can

lead to signiﬁcant improvements in ankle range of

motion, muscle strength, balance, self-reported func-

tion, quality of movement, daily activities, and upper

and lower limb function.

A range of intelligent technologies, particularly

those using machine learning, are increasingly be-

ing integrated into patient rehabilitation. Numerous

Zahwanie, R. H., Cheniti-Belcadhi, L. and Layouni, S.

Ontology-Based Framework for Personalized Home-Based Rehabilitation in Cerebral Palsy Care.

DOI: 10.5220/0013472700003932

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

In Proceedings of the 17th International Conference on Computer Supported Education (CSEDU 2025) - Volume 1, pages 705-716

ISBN: 978-989-758-746-7; ISSN: 2184-5026

705

studies highlight the potential of these technologies

in stroke rehabilitation and home care, respectively.

In (Mennella et al., 2023b) the author highlights the

role of artiﬁcial intelligence in remote monitoring and

intelligent assistance, particularly in activity recog-

nition and clinical status prediction. This research

(Fong et al., 2020) extends this discussion to the use

of intelligent robotics and machine learning in func-

tional capacity assessment and rehabilitation, partic-

ularly in the simulation of tasks in a suitable en-

vironment. These studies collectively highlight the

potential of intelligent technologies to improve the

efﬁciency and accessibility of patient rehabilitation.

Technological advances are revolutionising the reha-

bilitation of people with cerebral palsy. Robotics,

virtual reality, mobile applications, wearable sensors,

and non-invasive brain stimulation all hold promise

for improving movement, providing feedback, and

motivating users (as studies have shown). These ad-

vances can ultimately improve a person’s ability to

carry out daily activities and participate more fully in

life.

Therefore, a coherent conceptual framework and

knowledge representation are highly desirable and

necessary for the selection and recommendation of

appropriate rehabilitation exercises and recovery from

cerebral palsy. Several studies have proposed using

an ontology to represent and model domain knowl-

edge to overcome this problem. Ontology is a power-

ful tool for tackling semantic challenges and harmo-

nizing disparate vocabularies by providing a formal-

ized and structured representation of domain knowl-

edge. This knowledge is encapsulated in a way that

allows for clear deﬁnitions of concepts, relationships,

and rules within a particular domain.

In section 1, we review related work, examining

existing research and technological advancements in

the ﬁeld of cerebral palsy rehabilitation. In section

2, we focus on the architecture of the Ontological

Intelligent Framework for Rehabilitation. This sec-

tion outlines the system’s architecture and core com-

ponents. In section 3, we present a comprehensive

overview of the ontology used in our system, dis-

cussing its structure and the methodology behind its

development. Additionally, we focus on the Seman-

tic Web Rule Language (SWRL) rules applied within

the ontology to enhance its functionality and maintain

semantic coherence. In Section 4, we delve into the

implications of our ﬁndings, assess the effectiveness

of the proposed system, and outline potential direc-

tions for future research. Section 5 concludes the arti-

cle with a comprehensive summary of the key insights

and outcomes, highlighting their signiﬁcance.

2 RELATED WORK

Numerous studies have highlighted the need for rig-

orous monitoring of the development of children with

cerebral palsy throughout their rehabilitation home-

based exercise program. This research underlines the

importance of a personalized approach, adapted to the

speciﬁc needs of each child, in order to optimise ther-

apeutic results.

2.1 Intelligent Technologies Used in

Rehabilitation

Artiﬁcial intelligence, virtual reality, and wearable

devices enable more personalized and effective reha-

bilitation. Virtual reality immerses patients in simu-

lated environments to relearn movements, while ex-

oskeletons assist those with motor difﬁculties. Ma-

chine learning analyzes patient data to adapt rehabil-

itation programs in real time, optimizing outcomes.

Ontologies further enhance rehabilitation by structur-

ing and representing knowledge formally. These tech-

nological advances accelerate recovery and improve

the quality of life for individuals with disabilities or

recovering from illness.

The paper (Chu et al., 2022) discusses how smart

rehabilitation clinics are utilizing advanced technolo-

gies to improve patient care. The focus has shifted

towards a holistic approach, prioritizing patient well-

being and active participation for optimal results. The

article examines the deﬁning characteristics of these

clinics, including the integration of digital tools, au-

tomation, big data analysis, and patient monitoring

systems. These technologies have the potential to per-

sonalize and improve rehabilitation experiences. Ul-

timately, the paper presents a promising future for

rehabilitation clinics that embrace cutting-edge tech-

nologies and prioritize patient engagement to improve

overall quality of care and patient well-being.

Authors in (Lopes et al., 2021) emphasize the im-

portance of digital platforms in optimizing health-

care in pandemics like COVID-19. Scientiﬁc re-

search is crucial before deploying any digital prod-

ucts in this sector. The researchers also explore AI

in biomedicine, healthcare care, and medical educa-

tion, highlighting its potential for improved diagnos-

tics, personalized treatment plans, and enhanced edu-

cation.

The research (Senbekov et al., 2020) examines a

new rehabilitation system for schizophrenia patients

using virtual reality (VR) and serious games. The sys-

tem focuses on patient engagement and includes three

core modules: data integration, game design, and data

visualization. The research suggests integrating this

EKM 2025 - 8th Special Session on Educational Knowledge Management

706

system with traditional treatments to create a compre-

hensive rehabilitation program in healthcare facilities.

2.2 Rehabilitation-Based Home

Exercises in Healthcare

Chung et al. (Chung et al., 2019) discuss the appli-

cation of IoT in healthcare, with a particular focus on

exercise in cardiac rehabilitation at home. The sys-

tem comprises a wearable device, a smartphone ap-

plication, and a medical station to facilitate patient-

provider interaction, accurate heart rate measurement,

real-time exercise intensity recommendations, and ex-

ercise record tracking. The system has proven to be

effective in maintaining heart rate within prescribed

limits during exercise.

In addition, the main approach proposed in the

study (Triantafyllidis et al., 2018) is the development

of a computerized decision support system (DSS) for

home-based rehabilitation in cardiovascular disease

patients leverages rule-based logic to guide personal-

ized exercise programs. By focusing on heart rate and

movement accuracy, the system helps patients per-

form exercises effectively. Evaluations through simu-

lations and real-life studies demonstrate its efﬁcacy in

maintaining optimal heart rate zones during exercise.

In (Mennella et al., 2023a), the researcher evalu-

ates the use of deep learning-based systems to assess

rehabilitation exercises, focusing on human pose esti-

mation, movement classiﬁcation, and the detection of

compensatory movements. The systems discussed in

the review show promising results in accurately clas-

sifying exercise movements and providing real-time

feedback to patients during home-based rehabilitation

programs.

Therefore, AI-based systems in rehabilitation of-

fer real-time feedback, correct exercise execution,

detect compensatory movements, and improve out-

comes. Integration enhances patient engagement, en-

ables cost-effective monitoring, and supports HEP. AI

technologies accurately classify movements and de-

tect patterns, improving physical function and qual-

ity of life in various populations. Further, validation

studies in clinical settings are needed to assess their

effectiveness fully.

2.3 Ontology-Driven Rehabilitation

Frameworks

Existing ontologies in rehabilitation, such as the Re-

habilitation Treatment Speciﬁcation System (RTSS)

(Gibson et al., 2023) and the Ontology for Neurolog-

ical Rehabilitation (NeuRO) (Hier and Brint, 2020),

provide foundational models for rehabilitation pro-

cesses but lack speciﬁcity for cerebral palsy (CP) re-

habilitation. RTSS standardizes rehabilitation treat-

ments, including therapeutic exercises and modali-

ties, but does not cover CP-speciﬁc interventions like

spasticity management or neurodevelopmental facili-

tation. Similarly, NeuRO focuses on stroke and brain

injury rehabilitation, omitting pediatric CP subtypes

such as dyskinetic CP and adaptations for home-based

exercises. In contrast, our ontology explicitly ad-

dresses these gaps by structuring CP-speciﬁc con-

cepts (e.g., Motor Rehabilitation Techniques, Neu-

ropsychological Rehabilitation), integrating home-

based exercise programs (HEP) with severity-driven

rules (e.g., SWRL-based intensity adjustments), and

linking motor assessments to personalized interven-

tions, such as using ankle range-of-motion scores to

trigger targeted exercises.

2.4 Rehabilitation HEP in Cerebral

Palsy

In this section, we provide detailed information on ar-

ticles relating to the type of study, study design, and

intervention concerning HEP for cerebral palsy, de-

tailed in the table 1.

Table 1: HEP for cerebral palsy. F: Feasibility study, E:

Efﬁcacy study, C: Comparative study, RCT: Randomized

controlled trial, N-RCT: Non-Randomized controlled trial.

Authors Study

type

Intervention

(ling Chen

et al., 2014)

E Home-based Constraint-

Induced Therapy

(Racic et al.,

2014)

C Home-based habilitation

programme

(Demeke

et al., 2023)

F Home-based therapy

programs for children

with cerebral palsy

(Lorentzen

et al., 2015)

N-

RCT

Interactive home-based

training delivered

through the internet.

(Beckers

et al., 2017)

F Home-based occu-

pational therapy and

physiotherapy pro-

grammes

(Junior et al.,

2021)

F Home-based physiother-

apy programmes

(Goswami

et al., 2021)

(RCT)

Home-centered activity-

based therapy

(Macintosh

et al., 2020)

F Video game

(Ejaz et al.,

2024)

E Home-based intensive

bimanual training

Ontology-Based Framework for Personalized Home-Based Rehabilitation in Cerebral Palsy Care

707

To summarize, home-based rehabilitation treat-

ment programs, including various therapies such as

intelligent training, restraint therapy, and virtual re-

ality therapy, seem to ensure that children with cere-

bral palsy progress in their motor skills, useful abil-

ities, and exercises of daily living. These methods

offer potential rewards in terms of improved access,

cost-effectiveness, and comfort for patients and their

families. In-depth research is needed to optimize the

agreements, evaluate long-term results, and decide on

the most persuasive approaches for speciﬁc subsets

of children with cerebral palsy. By leveraging ad-

vanced methodologies stakeholders can reﬁne their

approaches, ensuring that interventions are not only

evidence-based but also highly adaptable to individ-

ual circumstances. This effort will pave the way for

more impactful outcomes in their rehabilitation.

3 ARCHITECTURE OF THE

ONTOLOGICAL INTELLIGENT

FRAMEWORK FOR

REHABILITATION

The proposed system architecture for home-based ex-

ercise programs (HEP) follows an interactive frame-

work. The user submits a request through the inter-

face, which is processed by the functional layer via

an API. This layer leverages predictive models and

personalized recommendations, generating a seman-

tically coherent response with support from the se-

mantic layer, which applies ontologies and inference

rules. The data layer supplies necessary information

for reasoning, and the ﬁnal response is delivered back

to the user, completing the interaction cycle.

Our architecture is composed of 4 layers (user

layer, functional layer, semantic layer, data layer) as

presented in the ﬁgure below:

Figure 1: Architecture of HEP framework.

• The user layer (UI Layer) is the point of inter-

action between the user and the system. It takes

the form of various interfaces, such as web or mo-

bile applications. Thanks to this layer, users can

not only formulate queries and consult the corre-

sponding results, but also personalize their expe-

rience by conﬁguring their preferences according

to their speciﬁc needs.

• The functional layer manages all the system’s

functionalities, focusing on a personalized user

experience. Using predictive models, it antici-

pates the user’s needs by analysing their histor-

ical data and current context. These predictions

form the basis for personalized recommendations

that are continually reﬁned as a function of the

user’s interactions with the system. In this way,

the functional layer ensures that the user’s prefer-

ences are monitored and analysed over time, mak-

ing it possible to improve the relevance of recom-

mendations over time and create an increasingly

tailored user experience.

• The semantic layer constitutes the cognitive heart

of the system, guaranteeing the coherence and

uniﬁed interpretation of information. It houses

a structured representation of the application do-

main’s knowledge, using ontological models to

deﬁne a common vocabulary and the relationships

between concepts. These models, combined with

inference and logic rules, enable the system to rea-

son about the data and draw relevant conclusions.

In this context, cerebral palsy rehabilitation, the

semantic layer can establish links between spe-

ciﬁc symptoms and probable diseases, facilitating

diagnosis and decision-making.

• The data layer contains a variety of clinical data

such as examination results, treatment history and

general medical information about the patient.

This layer thus provides the raw material essen-

tial for the other layers of the system to carry out

their treatments and generate relevant results.

This architecture offers a number of signiﬁcant ad-

vantages. It offers a high degree of customisation by

adapting its responses to the speciﬁc needs of each

user. Moreover, its ﬂexibility is ensured by the use

of ontologies, making it easy to extend the system to

new domains and integrate new knowledge. In ad-

dition, the reliability of the results is guaranteed by

rigorous inference rules, while the explicability of the

reasoning is reinforced by the transparency of the on-

tological models used, enabling a better understand-

ing of the system’s decision-making processes.

Participation to physical activity therapy by pa-

tients is seen as a crucial element in enhancing their

EKM 2025 - 8th Special Session on Educational Knowledge Management

708

overall health and wellness. It has been discovered

that home-based physical activity therapies, used in

homes or other settings, improve daily assessment

and tutoring and help patients take part in regular

physical exercise. Such interventions become much

more valuable when they may be used conveniently,

anytime, and unsupervised. This helps patients live

independently. As a result, intelligent, individual-

ized, durable, and adaptable computer systems for

exercise-based rehabilitation are needed to meet pa-

tients’ evolving needs and produce the desired health

results.

The goal is to develop a system that provides

patients with protected access to their personal pro-

ﬁle, enables them to monitor their progress, and re-

ceives personalized recommendations for HEP. It ex-

plores the application of predictive analytics in fol-

lowing and assessing patients’ rehabilitation progress

in a smart HEP system and develops and evaluates

video recommendation algorithms for HEP in an in-

telligent system. The approach consists of process-

ing patient data, such as medical history, recent exer-

cises, and progressive data, to generate personalized

exercise recommendations: The implementation of

a video recommendation algorithm in the intelligent

HEP for rehabilitation introduces an important new

feature to the system, offering a more interactive and

engaging rehabilitation experience for patients by giv-

ing them tailor-made exercise videos that match their

speciﬁc needs and abilities.

4 ONTOLOGICAL DESCRIPTION

FOR REHABILITATION

Ontologies are crucial in rehabilitation medicine, es-

pecially in robotics, patient education, and physio-

therapy (Palagin et al., 2023) (Spoladore et al., 2024).

They provide structured information about rehabil-

itation robots, aid researchers and experts, develop

multi-agent platforms for medical rehabilitation, as-

sist physiotherapists in managing patients’ evolution,

and facilitate a formal model of rehabilitation in-

terventions in computer-based decision support sys-

tems (Abdullah et al., 2022). Ontologies facilitate

the development of multi-agent, semantic technology-

supported rehabilitation interventions. They provide a

foundational framework for home rehabilitation, en-

abling knowledge integration and sharing across do-

mains. By deﬁning relationships between key con-

cepts such as patient status, therapy types, and feed-

back mechanisms, ontologies help create intelligent,

personalized environments that optimize therapy ef-

fectiveness. Moreover, they enhance interoperability

between different rehabilitation systems, ensuring a

seamless and coherent experience. Ontologies also

reduce experts’ workload by simplifying the develop-

ment of rehabilitation practices, which is essential for

maintaining patient engagement and improving out-

comes. Ultimately, they play a crucial role in sup-

porting a more effective and personalized approach to

home rehabilitation.

The purpose of this work is to provide a frame-

work for the ontology-based cerebral palsy rehabili-

tation system. Open Biomedical Ontologies (OBO)

Foundry, one of the most popular and detailed ap-

proaches for ontology construction, serves as the

foundation for the methodological approach used to

create, execute, and portray the ontology in this study

(Karray et al., 2021). The OBO Ontology recovery

serves as the foundation for the OBO Foundry, which

is a cooperative work within ontology researchers

who have previously committed to implementing a

wider range of norms outlining best practices for on-

tology development. These guidelines aim to guar-

antee a progressive enhancement of formal accuracy

and reliability in ontologies, in order to satisfy the in-

creasing needs of data and information integration in

the biomedical ﬁeld. In the following, we detail our

ontology according to the proposed methodology.

Cerebral palsy is a neurological disorder affect-

ing movement and coordination, often resulting from

brain damage before, during, or after birth. Rehabil-

itation plays a crucial role in treatment, focusing on

improving motor function, communication, and so-

cial participation. The ontology we propose aims to

support home-based rehabilitation for cerebral palsy

patients by integrating tools and treatments. Its de-

velopment begins with a requirements analysis phase,

identifying key concepts, attributes, relationships, and

axioms. In the design phase, a structured conceptual

model is created, progressively increasing in com-

plexity to ensure coherence and effectiveness.

4.1 Ontology Overview

The ontology can be developed using editors such as

Prot

e or OWLGrinder. In this work, Prot

e was

used for ontology creation. This allowed for the trans-

formation of knowledge about the required function-

alities of the proposed rehabilitation software for chil-

dren with cerebral palsy from an informal model into

a semi-formal model accessible to both domain ex-

perts and ontology developers. To enhance the on-

tology’s robustness and expressiveness, a comprehen-

sive glossary was created, detailing key concepts and

basic terminologies with precise deﬁnitions and ex-

plicit links to relevant resources.

Ontology-Based Framework for Personalized Home-Based Rehabilitation in Cerebral Palsy Care

709

The hierarchical structure of our model is ex-

plained in Figure 2. his structure organizes the key

components of the pediatric rehabilitation ontology,

enabling a comprehensive and systematic representa-

tion of the domain. In the following section, we will

explore each class within the hierarchical structure

of the pediatric rehabilitation ontology in greater de-

tail. Pediatric rehabilitation focuses on improving

the physical, cognitive, social, and emotional well-

being of children with disabilities or developmental

challenges. It aims to enhance their independence and

quality of life through targeted interventions and ther-

apies.

Figure 2: The hierarchical classes of the ontology.

This part in ﬁgure 3 of the ontology focuses on

Equipment and Technical Aids. It categorizes dif-

ferent types of aids into three subclasses: Technolog-

ical Aids, Speciﬁc Adaptations, and Mobility Aids.

• Technological Aids include various assistive tech-

nologies such as Environmental Control Systems,

Computer Assistance Technologies, Communica-

tion Devices, Adapted Digital Interfaces, and Per-

sonalized Learning Tools. These are designed to

enhance communication, learning, and environ-

mental interaction for children in rehabilitation.

• Speciﬁc Adaptations cover customized solutions

like Compensatory Technologies, Sensory Stim-

ulation Equipment, School Adaptations, and

Adapted Wheelchairs, tailored to meet the unique

needs of each child in educational and daily living

environments.

• Mobility Aids include Locomotion Aids, Or-

thoses and Appliances, and Positioning Systems,

which support physical mobility, posture, and

alignment, contributing to improved indepen-

dence and comfort.

This part of the ontology (ﬁgure 4) explores the In-

terdisciplinary Approach, emphasizing collabora-

tive methods for providing holistic care. It is divided

into two main subclasses: Family-Centered Approach

and Care Coordination.

Figure 3: The sub-classes of the Equip-

ment and technical aids class.

Figure 4: The sub-classes of the Interdisciplinary approach

class.

• Family-Centered Approach includes strategies

that prioritize the involvement and support of the

family in the rehabilitation process. It encom-

passes Parental Support Strategies, Family Educa-

tion, Comprehensive Support, and Resources and

Support Networks, aiming to empower families

with knowledge, tools, and networks necessary

for effective participation in their child’s care.

• Care Coordination focuses on the organized

and collaborative efforts among healthcare pro-

fessionals. This includes Communication Be-

tween Professionals, Multidisciplinary Consulta-

tion Meetings, Regular Monitoring and Adjust-

ment, and the creation of a Personalized Treat-

ment Plan. These elements ensure that care is in-

tegrated, consistently monitored, and adapted to

meet the evolving needs of the child.

This part of the ontology (ﬁgure 5) focuses on Motor

and Functional Therapy, outlining key components

of motor therapy. It is categorized into three main

subclasses: Specialized Interventions, Initial Motor

Assessment and Motor Rehabilitation Techniques.

EKM 2025 - 8th Special Session on Educational Knowledge Management

710

• Specialized Interventions include targeted ther-

apies such as Neurological Stimulation Ap-

proaches, Spasticity Management, Orthopedic

Rehabilitation, Motor Control Techniques, and

Muscle Tone Analysis. These interventions ad-

dress speciﬁc motor impairments and aim to en-

hance motor control and reduce spasticity.

• Initial Motor Assessment covers the foundational

evaluation processes like Analysis of Posture and

Body Alignment, Spasticity Assessment, Assess-

ment of Joint Amplitudes, and other diagnostic

measures crucial for creating therapy plans.

• Motor Rehabilitation Techniques consist of meth-

ods like Muscle Strengthening Techniques, Neu-

rodevelopmental Facilitation Therapy, Joint Mo-

bilization Techniques, Neurological Physiother-

apy, and Muscle Relaxation Techniques. These

techniques are designed to improve motor func-

tion, ﬂexibility, and overall physical well-being.

Figure 5: The sub-classes of the Mo-

tor and Functional Therapy class.

This part focuses on the subdomain of Neuropsy-

chological Rehabilitation, a critical area within pe-

diatric rehabilitation that focuses on the intersection

of neurological and psychological care. It aims to ad-

dress the cognitive, emotional, and behavioral chal-

lenges faced by children with neurological conditions

or developmental disorders. It is subdivided into cat-

egories such as Psychological Support, Cognitive As-

sessment, and Cognitive Interventions.

• Psychological Support encompasses therapies

such as Behavioral and Emotional Therapies,

Stress and Adaptation Management, Family Sup-

port, and targeted support like Psychological Sup-

port for the Child, all aimed at addressing emo-

tional well-being and family dynamics.

• Cognitive Assessment includes methods to eval-

uate various cognitive functions, such as Analy-

sis of Attentional Abilities, Assessment of Cog-

nitive Functions, Memory and Learning Assess-

ment, and Targeted Cognitive Stimulation, en-

abling precise diagnostics of cognitive challenges.

• Cognitive Interventions involves strategies for re-

habilitation, including Cognitive Compensation

Strategies, Management of Associated Disorders,

Cognitive Rehabilitation Techniques, and Behav-

ioral Therapies, all focused on improving cogni-

tive and functional abilities. a critical area within

pediatric rehabilitation that focuses on the inter-

section of neurological and psychological care.

Figure 6: The sub-classes of the Neuropsychologi-

cal rehabilitation class.

This ontology represents the structure of Occupa-

tional Therapy within the broader domain of Pedi-

atric Rehabilitation. It is designed to support chil-

dren in achieving greater independence and improv-

ing their quality of life through functional and adap-

tive interventions tailored to their unique needs. The

ontology organizes this domain into key concepts and

categories that reﬂect the diverse aspects of occupa-

tional therapy.

• Functional Assessment focuses on evaluating the

child’s capabilities through methods such as Anal-

ysis of Home and School Environment, Assess-

ment of Activities of Daily Living, Analysis of

Fine Motor Skills, and Evaluation of Autonomy,

which together provide a comprehensive under-

standing of the child’s functional abilities and ar-

eas for improvement.

• Adaptive Interventions includes practical strate-

gies to improve daily functioning, such as Com-

pensation Techniques, Learning Daily Activities,

Ontology-Based Framework for Personalized Home-Based Rehabilitation in Cerebral Palsy Care

711

Adaptation of Tools and Equipment, Behavioral

Adaptation Strategies, and Development of Fine

Motor Skills, with an emphasis on personalized

tools and methods.

• Environmental Adaptations aims to modify the

child’s surroundings to better suit their needs.

This includes Sensory Compensation Strategies,

Adapting Furniture, Accessibility Recommenda-

tions, and Design of Personalized Technical Aids,

ensuring the environment facilitates the child’s re-

habilitation and independence.

Figure 7: The sub-classes of the Occupational Therapy

class.

This part, shown in ﬁgure 8 illustrates the Speech

Therapy domain within Pediatric Rehabilitation.

Speech therapy is a critical discipline aimed at ad-

dressing communication, language, and swallowing

difﬁculties in children, empowering them to interact

effectively with their environment and enhance their

quality of life. The ontology organizes this domain

into speciﬁc categories that reﬂect the comprehensive

nature of speech therapy interventions.

• Communication Assessment encompasses eval-

uating various aspects of communication, such

as Analysis of Orofacial Functions, Swallowing

Assessment, Analysis of Expressive and Recep-

tive Language, and Assessment of Communica-

tion Skills, while also exploring Alternative Com-

munication Techniques to support children with

speciﬁc challenges.

• Therapeutic Interventions targets the develop-

ment and enhancement of communication abili-

ties through methods like Development of Lan-

guage Skills, Swallowing Therapy, Articulation

Rehabilitation, and Language Stimulation Tech-

niques, often integrating Use of Technological

Devices to assist in therapy.

• Communication Tools provides resources to sup-

port speech therapy, including Voice Assistance

Technologies, Adapted Communication Systems,

and Augmentative and Alternative Communica-

tion methods, ensuring tailored support for each

child’s unique needs.

Figure 8: The sub-classes of the Speech Therapy class.

4.2 SWRL Rules

In addition, to enhance the ontology’s reasoning ca-

pabilities, a set of SWRL rules has been meticulously

developed.

These rules, organized in the form of a structured

list, enable new knowledge to be inferred from ex-

isting information within the ontology. This rigorous

approach guarantees a formal and coherent represen-

tation of domain knowledge, enabling the ontology to

be better understood and used more effectively.

Query 1:

Patient(?p) ∧ hasAssessment(?p, ?a) ∧

hasAssessmentType(?a, ”Functional”) ∧

hasAssessmentScore(?a, ?score) ∧

lessThan(?score, 40.0) →

requiresAdaptation(?p, Environmental Adaptation)

This logical rule is designed to automate decision-

making processes within the ontology, ensuring that

patients with limited functional abilities (as indicated

by a low assessment score) are identiﬁed for environ-

mental modiﬁcations to support their rehabilitation

needs. It incorporates both semantic reasoning and

threshold-based logic for personalized recommenda-

tions.

EKM 2025 - 8th Special Session on Educational Knowledge Management

712

Table 2: Object Properties.

Data properties Deﬁnition

PatientID Represents the unique identi-

ﬁer assigned to a patient.

Condition Describes the medical condi-

tion or diagnosis of the pa-

tient.

AdmissionDate Indicates the date on which

the patient was admitted to

the program.

DischargeDate Speciﬁes the date on which

the patient was discharged

from the program.

ProgressScore Represents the progress

score of the patient, mea-

sured on a scale of 0.0 to

10.0.

AssessmentDate Speciﬁes the date when the

assessment was conducted.

AssessmentScore Represents the score ob-

tained in the assessment, on

a scale of 0.0 to 100.0.

AssessmentType Indicates the type or cate-

gory of the assessment.

Recommendation Provides recommendations

based on the assessment.

SessionDate Speciﬁes the date when the

therapy session occurred.

SessionDuration Represents the duration of

the therapy session, in min-

utes.

TherapyGoal Describes the goal or objec-

tive of the therapy session.

ProgressNotes Provides notes about the

progress made during the

session.

Frequency Indicates the frequency of

therapy sessions (e.g., ses-

sions per week).

IntensityLevel Speciﬁes the intensity level

of the therapy session (Low,

Medium, or High).

EquipmentID Represents the unique iden-

tiﬁer assigned to a piece of

equipment.

MaintenanceDate Speciﬁes the date of the last

maintenance of the equip-

ment.

Speciﬁcation Describes the technical spec-

iﬁcations of the equipment.

AvailabilityStatus Indicates whether the equip-

ment is currently available

(True) or not (False).

Table 3: Data Properties.

Object properties Deﬁnition

hasTherapist Associates a patient with

a therapist responsible for

their rehabilitation.

assignedTo Links a therapist to a speciﬁc

rehabilitation program.

participatesIn Indicates that a patient par-

ticipates in a given therapy

session.

supervises Speciﬁes that a therapist su-

pervises a therapy session.

requiresEquipment Indicates that a therapy ses-

sion requires speciﬁc techni-

cal equipment.

hasAssessment Links a patient to an as-

sessment that evaluates their

abilities or needs.

follows Indicates that a patient fol-

lows a speciﬁc treatment

plan.

collaboratesWith Represents collaboration be-

tween two therapists.

supportedBy Associates a patient with

a family member providing

support during their rehabil-

itation.

usesTool Speciﬁes that speech therapy

uses a communication tool.

appliesTechnique Indicates that motor therapy

applies a speciﬁc rehabilita-

tion technique.

requiresAdaptation Indicates that a patient re-

quires environmental adapta-

tions for their rehabilitation.

utilizesAid Speciﬁes that a patient uses a

mobility aid during therapy.

hasIntervention Indicates that a therapy ses-

sion includes a speciﬁc ther-

apeutic intervention.

conductsAssessment Speciﬁes that a therapist con-

ducts a given assessment.

Query 2:

Patient(?p) ∧ hasProgressScore(?p, ?score) ∧

lessThan(?score, 6.0) =⇒

requiresAssessment(?p, Cognitive Assessment)

This rule identiﬁes patients who need a cognitive

assessment based on their progress score. If a Patient

(?p) has a Progress Score (?score) below the thresh-

old of 6.0, the system infers that the patient requires

a Cognitive Assessment. This ensures that cogni-

tive evaluations are prioritized for individuals show-

ing signs of limited progress, potentially highlighting

areas requiring intervention.

Ontology-Based Framework for Personalized Home-Based Rehabilitation in Cerebral Palsy Care

713

Query 3:

Patient(?p) ∧ hasAssessment(?p, ?a) ∧

hasAssessmentScore(?a, ?score) ∧

lessThan(?score, 50.0) ∧

hasTherapyType(?p, Motor Therapy) =⇒

hasIntensityLevel(?p, ”High”)

This rule focuses on adjusting the intensity of mo-

tor therapy based on assessment scores. If a Patient

(?p) has undergone an Assessment (?a) with an As-

sessment Score (?score) below 50.0 and is already

undergoing Motor Therapy, the system recommends

increasing the therapy’s intensity by setting the Inten-

sity Level of the patient to ”High”. This rule supports

personalized treatment plans by dynamically adjust-

ing motor therapy intensity based on a patient’s per-

formance and needs.

5 DISCUSSION

The development and implementation of our

ontology-based framework for cerebral palsy reha-

bilitation presents several signiﬁcant contributions

and implications for home-based exercise programs

(HEP). Our ﬁndings highlight both the potential

beneﬁts and challenges in implementing such a

semantic framework for rehabilitation. The proposed

ontological framework demonstrates several key

advantages for cerebral palsy rehabilitation. The

hierarchical structure of our ontology successfully

integrates diverse aspects of rehabilitation, from

equipment and technical aids to specialized therapies.

This comprehensive approach ensures that all crucial

elements of rehabilitation are systematically repre-

sented and interlinked. By adopting OBO Foundry

principles, our ontology promotes standardization

and interoperability with existing biomedical ontolo-

gies. This compatibility enhances the potential for

knowledge sharing across different healthcare sys-

tems and rehabilitation platforms. Furthermore, the

framework’s ability to capture complex relationships

between patient characteristics, rehabilitation tech-

niques, and outcomes enables highly personalized

treatment recommendations, which is particularly

valuable for home-based exercise programs where

individualized guidance is crucial.

The implementation of our framework has several

important implications for clinical practice. The se-

mantic rules and relationships deﬁned in our ontol-

ogy provide a robust foundation for clinical decision

support systems, allowing healthcare providers to

leverage this structured knowledge to make more in-

formed decisions about rehabilitation protocols. The

framework’s comprehensive modeling of rehabilita-

tion concepts enables better remote monitoring of pa-

tient progress, which is essential for home-based ex-

ercise programs and particularly relevant given the

increasing importance of telerehabilitation solutions.

Additionally, the standardized representation of re-

habilitation concepts and procedures helps maintain

consistency in treatment delivery and assessment,

even in remote settings. Several technical aspects

warrant discussion in our implementation. While our

ontology successfully models the complex domain

of cerebral palsy rehabilitation, considerations must

be made for scaling the system to handle larger pa-

tient populations and additional rehabilitation proto-

cols. The implementation of semantic rules through

SWRL presents challenges in terms of computational

efﬁciency and real-time processing, suggesting that

future work should focus on optimizing rule execu-

tion for time-sensitive applications. Moreover, the

comprehensive nature of our ontology raises impor-

tant considerations about data privacy and security,

particularly when handling sensitive patient informa-

tion in home-based settings.

Based on our ﬁndings, several areas merit further

investigation. Future work should explore the inte-

gration of machine learning algorithms with our on-

tological framework to enhance the adaptive capabil-

ities of rehabilitation recommendations. The frame-

work should be expanded to accommodate cultural

and linguistic variations in rehabilitation practices,

ensuring broader applicability across different health-

care systems. Large-scale clinical validation studies

are needed to assess the framework’s effectiveness in

improving rehabilitation outcomes compared to tradi-

tional approaches. Additionally, the development of

mobile applications that can seamlessly interact with

the ontological framework would enhance accessibil-

ity and user engagement in home-based settings.

Several limitations of our current work should be

acknowledged. The current validation of the ontology

has been primarily theoretical and would beneﬁt from

extensive real-world testing in clinical settings. The

implementation of our framework requires signiﬁcant

technical infrastructure, which may limit its adoption

in resource-constrained environments. While com-

prehensive for cerebral palsy rehabilitation, the ontol-

ogy may require substantial modiﬁcations for appli-

cation to other rehabilitation domains.

The proposed ontology-based framework repre-

sents a signiﬁcant step forward in structuring and de-

livering home-based rehabilitation programs for cere-

bral palsy patients. However, its successful imple-

mentation will require careful consideration of both

technical and practical challenges, along with contin-

EKM 2025 - 8th Special Session on Educational Knowledge Management

714

ued reﬁnement based on clinical feedback and tech-

nological advancements.

6 CONCLUSION

In conclusion, our ontology for cerebral palsy reha-

bilitation marks a signiﬁcant advancement in home-

based exercise programs. By integrating diverse re-

habilitation components within a hierarchical struc-

ture and adhering to OBO Foundry principles, it en-

sures standardization and interoperability with ex-

isting biomedical ontologies. This comprehensive

framework facilitates personalized treatment recom-

mendations and supports clinical decision-making

through well-deﬁned semantic rules and relation-

ships.

The implementation has shown great potential in

enhancing remote monitoring and ensuring consis-

tency in treatment delivery, which is especially vital

for home-based rehabilitation. However, challenges

remain, such as scalability and optimizing rule execu-

tion for time-sensitive applications, indicating areas

for future improvement.

Future research should prioritize integrating ma-

chine learning algorithms to improve adaptability, ad-

dressing cultural and linguistic diversity, and validat-

ing the framework through large-scale clinical stud-

ies. Developing accessible mobile applications could

enhance user engagement in home-based settings.

While challenges such as real-world testing and tech-

nical infrastructure persist, the framework establishes

a solid foundation for advancing rehabilitation prac-

tices.

REFERENCES

Abdullah, Z. H., Zakaria, L. Q., Ismail, W., and Ismail,

S. (2022). Developing an exercise games ontology

towards standard rehabilitation clinical terms. In

2022 Euro-Asia Conference on Frontiers of Computer

Science and Information Technology (FCSIT), pages

173–176. IEEE.

Beckers, L., Schnackers, M., Janssen-Potten, Y., Kleijnen,

J., and Steenbergen, B. (2017). Feasibility and ef-

fect of home-based therapy programmes for children

with cerebral palsy: a protocol for a systematic review.

BMJ Open, 7(2).

Chen, K., Ren, Y., Gaebler-Spira, D., and Zhang, L.-Q.

(2014). Home-based tele-assisted robotic rehabili-

tation of joint impairments in children with cerebral

palsy. In 36th Annual International Conference of the

IEEE Engineering in Medicine and Biology Society.

Chu, E. C.-P., Yun, S. M. H., Hau, K. H. Y., and Piong,

S. Z. (2022). Smart rehabilitation clinic. Journal of

Contemporary Chiropractic, 5:7–12.

Chung, H., Lee, H., Kim, C., Hong, S., and Lee, J. (2019).

Patient-provider interaction system for efﬁcient home-

based cardiac rehabilitation exercise. IEEE Access,

7:14611–14622.

Demeke, Z. D., Assefa, Y. A., Abich, Y., and Chala, M. B.

(2023). Home-based therapy and its determinants for

children with cerebral palsy, exploration of parents’

and physiotherapists’ perspective, a qualitative study,

ethiopia. PLoS One, 18(2).

Ejaz, F., Tanveer, F., Shoukat, F., Fatima, N., and Ahmad,

A. (2024). Effectiveness of routine physical therapy

with or without home-based intensive bimanual train-

ing on clinical outcomes in cerebral palsy children: a

randomised controlled trial. Physiotherapy Quarterly,

32(1):78–83.

Fong, J., Ocampo, R., Gross, D. P., and Tavakoli, M. (2020).

Intelligent robotics incorporating machine learning al-

gorithms for improving functional capacity evaluation

and occupational rehabilitation. Journal of Occupa-

tional Rehabilitation, 30:362–370.

Ge, Y., Zhao, W., Zhang, L., Zhao, X., Shu, X., Li, J., Qiao,

L., Liu, Y., and Wang, H. (2024). Home physical ther-

apy versus telerehabilitation in improving motor func-

tion and quality of life in parkinson’s disease: a ran-

domized controlled trial. BMC Geriatrics, 24(1):968.

Gibson, J., Sampford, J., Myers-Ingram, R., and Jones,

G. D. (2023). Embedding the rehabilitation treatment

speciﬁcation system (rtss) into clinical practice: an

evaluation of a pilot teaching programme. BMC Med-

ical Education, 23(1):85.

Goswami, J. N., Sankhyan, N., and Singhi, P. (2021). Add-

on home-centered activity-based therapy vs conven-

tional physiotherapy in improving walking ability at

6-months in children with diplegic cerebral palsy:

A randomized controlled trial. Indian Pediatrics,

58:826–832.

Hier, D. B. and Brint, S. U. (2020). A neuro-ontology for

the neurological examination. BMC Medical Infor-

matics and Decision Making, 20:1–9.

Ibeneme, S. C., Chimereze, N. R., Fortwengel, G., John,

J. N., Ibeneme, G. C., and Nwosu, A. O. (2024). Ef-

fectiveness of home-based exercise programme in im-

proving function, activity and participation for chil-

dren with cerebral palsy: a systematic review and

meta-analysis.

Junior, P. R. F., Souza, P. P., dos Reis, K. K. M., and Filoni,

E. (2021). Home-based physiotherapy programmes

for individuals with neurological diseases: systematic

review. Fisioterapia em Movimento, 32.

Karray, M., Otte, N., Rai, R., Ameri, F., Kulvatunyou, B.,

Smith, B., Kiritsis, D., Will, C., Arista, R., et al.

(2021). The industrial ontologies foundry (iof) per-

spectives.

Komatireddy, R., Chokshi, A., Basnett, J., Casale, M.,

Goble, D., and Shubert, T. (2014). Quality and quan-

tity of rehabilitation exercises delivered by a 3-d mo-

tion controlled camera: A pilot study. International

Journal of Physical Medicine & Rehabilitation.

ling Chen, C., chung Lin, K., ju Kang, L., and yi Wu, C.

(2014). Potential predictors of functional outcomes

Ontology-Based Framework for Personalized Home-Based Rehabilitation in Cerebral Palsy Care

715

after home-based constraint-induced therapy for chil-

dren with cerebral palsy. The American Journal of

Occupational Therapy, 68.

Lopes, R. P., Barroso, B., Deusdado, L., Novo, A.,

Guimar

aes, M., Teixeira, J. P., and Leit

ao, P. (2021).

Digital technologies for innovative mental health re-

habilitation. Electronics, 10:2260.

Lorentzen, J., Greve, L. Z., Kliim-Due, M., Rasmussen, B.,

Bilde, P. E., and Nielsen, J. B. (2015). Twenty weeks

of home-based interactive training of children with

cerebral palsy improves functional abilities. BMC

Neurology, 15:1–12.

Macintosh, A., Vignais, N., Desailly, E., Biddiss, E., and

Vigneron, V. (2020). A classiﬁcation and calibration

procedure for gesture speciﬁc home-based therapy ex-

ercise in young people with cerebral palsy. IEEE

Transactions on Neural Systems and Rehabilitation

Engineering, 29:144–155.

Mennella, C., Maniscalco, U., Pietro, G. D., and Esposito,

M. (2023a). A deep learning system to monitor and

assess rehabilitation exercises in home-based remote

and unsupervised conditions. Computers in Biology

and Medicine, 166:107485.

Mennella, C., Maniscalco, U., Pietro, G. D., and Esposito,

M. (2023b). The role of artiﬁcial intelligence in future

rehabilitation services: a systematic literature review.

IEEE Access, 11:11024–11043.

O’Neill, D. and Forman, D. E. (2020). The importance of

physical function as a clinical outcome: assessment

and enhancement. Clinical Cardiology, 43:108–117.

Palagin, O., Kaverinsky, V., Petrenko, M., and Malakhov,

K. (2023). Digital health systems: Ontology-based

universal dialog service for hybrid e-rehabilitation ac-

tivities support. In 2023 IEEE 12th International

Conference on Intelligent Data Acquisition and Ad-

vanced Computing Systems: Technology and Applica-

tions (IDAACS), volume 1, pages 84–89. IEEE.

Racic, M., Kusmuk, S., and Krstovic-Spremo, V. (2014).

Effectiveness of the home-based habilitation pro-

gramme for children with cerebral palsy.

Senbekov, M., Saliev, T., Bukeyeva, Z., Almabayeva, A.,

Zhanaliyeva, M., Aitenova, N., Toishibekov, Y., and

Fakhradiyev, I. (2020). The recent progress and ap-

plications of digital technologies in healthcare: a re-

view. International Journal of Telemedicine and Ap-

plications.

Spoladore, D., Colombo, V., Fumagalli, A., Tosi, M.,

Lorenzini, E. C., and Sacco, M. (2024). An

ontology-based decision support system for tailored

clinical nutrition recommendations for patients with

chronic obstructive pulmonary disease: Development

and acceptability study. JMIR Medical Informatics,

12:e50980.

Triantafyllidis, A., Filos, D., Buys, R., Claes, J., Cor-

nelissen, V., Kouidi, E., Chatzitoﬁs, A., Zarpalas, D.,

Daras, P., Walsh, D., et al. (2018). Computerized

decision support for beneﬁcial home-based exercise

rehabilitation in patients with cardiovascular disease.

Computer Methods and Programs in Biomedicine,

162.

EKM 2025 - 8th Special Session on Educational Knowledge Management

716