=Paper=
{{Paper
|id=Vol-3806/S_6_Malakhov_Semykopna
|storemode=property
|title=
Integrating Hybrid Cloud Solutions in Telerehabilitation
|pdfUrl=https://ceur-ws.org/Vol-3806/S_6_Malakhov_Semykopna.pdf
|volume=Vol-3806
|authors=Kyrylo Malakhov,Tetyana Semykopna
|dblpUrl=https://dblp.org/rec/conf/ukrprog/MalakhovS24
}}
==
Integrating Hybrid Cloud Solutions in Telerehabilitation
==
https://ceur-ws.org/Vol-3806/S_6_Malakhov_Semykopna.pdf
Integrating Hybrid Cloud Solutions in Telerehabilitation
Kyrylo Malakhov1,∗,†, Tetyana Semykopna1,†
1 Glushkov Institute of Cybernetics of the National Academy of Sciences of Ukraine, 40 Glushkov ave., Kyiv, 03187,
Ukraine
Abstract
The digitalization of scientific research has significantly advanced with the use of information
and communication technologies, particularly in the field of physical and rehabilitation medicine
and its telerehabilitation branch. A hybrid cloud platform for telerehabilitation medicine is
implemented as a component-based collection of services, including a medical information
system for rehabilitation, a dialog information and reference system (MedRehabBot), a predictive
and analytical system for assessing rehabilitation process effectiveness, and services for
optimizing rehabilitation process models. These services operate within an ontology-driven,
service-oriented architecture. One of the key advantages of this type of architecture is its support
for experimental design systems, where the design process of the target architecture is
accompanied by scientific research. This paper examines the overall functional architecture of
the platform (and its technical requirements) in the form of three interacting subsystems:
medical-rehabilitation, information-analytical, and telerehabilitation. The architectural and
technological organization of the platform is developed using a model that implements an
advanced concept of an automated scientific research workstation. The main practical
achievement is the implementation and deployment of this architectural and technological
organization of the platform, which opens new opportunities for telerehabilitation in medicine.
Keywords
Hybrid cloud platform, Cloud computing, telerehabilitation, automated research workstation1
1. Introduction
The advancement of modern technologies significantly impacts intellectual activities,
particularly in the realm of research and development. In this context, a new class of
information systems has emerged—Research and Development Workstation Environment
(RDWE)—which implements an advanced concept of an automated workstation (AWS) for
ongoing research and associated intelligent information technologies. These systems and
concepts encompass the primary stages of the research and development lifecycle: from the
semantic analysis of information materials across various subject domains to the
development of innovative proposals' constructive features. A distinguishing feature of
RDWE systems is their ability to adapt (problem-oriented) to different types of scientific
14th International Scientific and Practical Conference from Programming UkrPROG’2024, May 14-15, 2024, Kyiv,
Ukraine
∗ Corresponding author.
† These authors contributed equally.
malakhovks@nas.gov.ua (K. Malakhov); semtv@ukr.net (T. Semykopna)
0000-0003-3223-9844 (K. Malakhov); 0000-0002-4116-0567 (T. Semykopna)
© 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
activities through the integration of various functional services and the ability to add new
ones within a hybrid cloud environment (platform).
A detailed analysis of the formal model, fundamental principles, and requirements for
developing RDWE-class information systems is provided in [1, 2]. Among the most
impressive examples of modern RDWE systems are the automated interactive system
OntoChatGPT [3], developed using advanced computational linguistics technologies such as
GPT-4 by OpenAI, ontology engineering support services and natural language
understanding KEn [4], UkrVectōrēs [5]. The OntoChatGPT system not only expands the
possibilities for intuitive human-machine interaction but also serves as a strategic tool in
the RDWE context, promoting the development of innovative information systems for
scientific research. A detailed overview of the OntoChatGPT RDWE system and information
about its evolution can be found in [3, 6].
In early 2022, a research team from the Institute of Cybernetics (including the authors
of this article) led by scientific supervisor Petro Stetsyk became one of the winners of the
"Science for Security and Sustainable Development of Ukraine" competition, organized by
the National Research Foundation of Ukraine (NRFU). The new project was titled
"Development of the Cloud-Based Platform for Patient-Centered Telerehabilitation of
Oncology Patients with Mathematical-Related Modeling" [7]. The project is dedicated to
developing a hybrid cloud platform and creating on its basis an information technology for
telerehabilitation of cancer patients, serving a wide range of specialists in Physical Medicine
and Rehabilitation (PM&R) in the "Telerehabilitation of Cancer Patients" sector. The
project's goal is to create a promising cloud platform for the telerehabilitation of cancer
patients based on the application of mathematical methods of system analysis, modeling,
and optimization. The choice of this direction is driven by the accelerated growth rate of
cancer patients in Ukraine [8]. The main idea of the approach to creating telerehabilitation
tools is the combination of artificial intelligence methods and means with mathematical
methods for solving complex problems in the chosen application domain. The project's main
tasks include:
Development of a hybrid cloud platform for telerehabilitation medicine (HCP-TM)
(with all its components, including services, platform, infrastructure) and its
architectural and technological organization (considering all the features of the
subject domain and the modern approach to digitalizing scientific research).
Development of an information-analytical subsystem (IAS) of the HCP-TM for
processing data generated in spatially distributed network sources, continuously
interacting with all profile specialists, participants in the rehabilitation process,
using interactive intelligent methods and tools implemented in the cloud platform.
The IAS is represented, in particular, by an AWS in the field of computational
linguistics with a specialized set of services and problem-oriented dataset.
This work examines the overall functional architecture of the hybrid HCP-TM (and its
technical requirements) in the form of three interacting subsystems: medical-rehabilitation
(MRS), information-analytical (IAS), and telerehabilitation (TRS). The architectural and
technological organization of the hybrid HCP-TM is also developed using the RDWE system
model.
2. Related Work
The field of telerehabilitation has experienced significant advancements through the
integration of cloud computing, the Internet of Medical Things (IoMT), remote patient
monitoring systems, hospital information systems, multimodal chatbots, and large language
models (LLMs). Additionally, ontology-related technologies are being increasingly used to
enhance these systems, providing structured and interoperable data frameworks crucial for
efficient telerehabilitation.
Cloud Computing in Telerehabilitation. Cloud computing is a critical enabler for
telerehabilitation, offering scalable storage and computational power necessary for
handling large volumes of data. The cloud's ability to support advanced data analytics and
machine learning models is crucial for personalized patient care. Studies demonstrate that
cloud computing facilitates real-time updates, remote access to patient records, and
seamless data integration, which are vital for efficient healthcare delivery [9, 10].
Internet of Medical Things (IoMT). IoMT integrates various wearable devices, sensors,
and mobile applications to monitor patients' health metrics continuously. This ecosystem
supports real-time data collection and transmission, enabling remote monitoring and
timely interventions. IoMT systems are particularly effective in chronic disease
management and rehabilitation, where continuous monitoring is essential for tracking
patient progress and adjusting treatment plans. Research highlights that IoMT significantly
improves patient outcomes by providing accurate and real-time health data [11, 12].
Remote Patient Monitoring (RPM). RPM systems leverage IoMT and cloud computing to
provide comprehensive care outside traditional clinical settings. These systems collect data
from various IoMT devices and use cloud-based analytics to generate actionable insights for
healthcare providers. RPM enhances patient engagement and adherence to rehabilitation
protocols by allowing active participation in their own care. Studies indicate that RPM
reduces hospital readmissions, improves chronic disease management, and enhances
patient satisfaction [13, 14].
Hospital Information Systems (HIS). Hospital Information Systems integrated with cloud
and IoMT technologies create a seamless digital workstation environment for healthcare
professionals. These systems streamline patient data management, clinical workflows, and
administrative tasks, facilitating efficient information sharing across departments. HIS
platforms typically include electronic health records (EHRs), clinical decision support
systems (CDSS), and telemedicine capabilities, which are essential for effective
telerehabilitation services. Implementation of HIS in telerehabilitation has been shown to
improve care quality, reduce operational costs, and enhance patient safety [15].
Multimodal Chatbots and Large Language Models in Telerehabilitation. Multimodal
chatbots represent the next frontier in enhancing patient interaction and care in
telerehabilitation. These chatbots leverage multiple modes of input, such as text, voice, and
images, to provide more interactive and intuitive patient support. Recent developments in
multimodal AI have led to chatbots that can understand and respond using various data
types. Advanced chatbots can process both textual and visual data to generate contextually
appropriate responses, improving patient engagement and support [15, 16].
Large language models (LLMs) like GPT-4 have demonstrated significant potential in the
field of telerehabilitation by providing sophisticated natural language understanding and
generation capabilities. These models can facilitate various aspects of telerehabilitation,
including personalized patient interactions, automated documentation, and decision
support. LLMs enhance the efficiency of telehealth services by enabling more accurate and
contextually relevant responses, thereby improving the quality of remote patient
care [17, 18].
One key application of LLMs in telerehabilitation is their ability to assist in remote
consultations. LLMs can help clinicians by transcribing and summarizing patient
interactions, generating reports, and providing evidence-based recommendations.
Additionally, these models can support patients directly by answering their queries,
providing exercise instructions, and offering motivational support [19].
LLMs also contribute significantly to data analysis and interpretation. They can analyze
patient data collected from IoMT devices to identify patterns and anomalies, enabling early
detection of potential health issues and timely interventions. This capability is particularly
valuable in chronic disease management and post-operative rehabilitation, where
continuous monitoring and prompt response are critical [20].
Ontology-Related Technologies in Telerehabilitation. Ontology-related technologies [21,
22, 23] provide a structured framework for data management and interoperability in
telerehabilitation [24, 25]. These technologies enable the integration and sharing of data
across different systems and applications, enhancing the effectiveness of telerehabilitation
services. Ontologies facilitate semantic data representation, allowing for better
understanding and utilization of data by both machines and humans. This is particularly
important in complex healthcare environments where data from various sources need to be
combined and analyzed.
Ontologies in telerehabilitation are used to standardize terminologies and data models,
ensuring that different systems can communicate effectively [26]. This standardization
enhances data interoperability and supports advanced data analytics and decision-making
processes. For example, by employing ontologies, healthcare providers can integrate
patient data from electronic health records (EHRs), wearable devices, and clinical
databases, leading to a more comprehensive view of the patient’s health status. Research
shows that the use of ontologies in healthcare can improve data quality, support
personalized medicine, and facilitate the development of intelligent healthcare applications
[27].
One of the significant benefits of using ontologies in telerehabilitation is the ability to
create detailed and precise rehabilitation protocols that can be easily shared and
understood by various stakeholders. These protocols can be tailored to individual patient
needs, ensuring that each patient receives personalized care. Ontology-based systems [28]
can also enhance the capabilities of decision support systems by providing a robust
framework for reasoning about patient data. This can lead to more accurate diagnoses,
better treatment plans, and improved patient outcomes [29].
Moreover, ontologies can support the automation of administrative and clinical
workflows in telerehabilitation [30]. By using ontology-driven systems, healthcare
providers can automate tasks such as appointment scheduling, resource allocation, and
patient follow-ups. This not only improves efficiency but also reduces the likelihood of
human errors. For instance, an ontology-based system can automatically generate
reminders for patients about their rehabilitation exercises or upcoming telehealth sessions,
ensuring better adherence to treatment plans. Studies have demonstrated that ontology-
driven approaches can significantly streamline healthcare operations and improve the
overall quality of care [30].
The integration of cloud computing, IoMT, RPM, HIS, multimodal chatbots, LLMs, and
ontology-related technologies forms a comprehensive digital workstation environment that
revolutionizes telerehabilitation. These technologies enhance data management, patient
monitoring, and clinical decision-making, leading to improved patient outcomes and more
efficient healthcare delivery. Ongoing research and development are expected to further
optimize these systems, making telerehabilitation more accessible and effective.
3. Hybrid Cloud Platform for Telerehabilitation Medicine
The development of the Hybrid Cloud Platform for Telerehabilitation Medicine (HCP-TM)
has been driven by the need to enhance the efficiency and effectiveness of telerehabilitation
services, particularly for oncology patients. This platform integrates artificial intelligence
(AI) with precise mathematical methods to optimize rehabilitation methodologies and the
entire telerehabilitation process. The combination of these technologies aims to provide
reliable patient assessments, effective intervention strategies, optimal rehabilitation
pathways, and accurate prognostications. The primary technical requirements for creating
the HCP-TM and its underlying information technology for oncology patient
telerehabilitation are outlined comprehensively. These requirements ensure the system's
robustness, scalability, and efficiency in managing rehabilitation processes. The detailed
technical requirements can be accessed via the provided references [31].
Currently, there is a tendency to intensify scientific research both at the intersection of
different subject disciplines (interdisciplinary research) and in convergence clusters
(transdisciplinary research). To support these studies, important factors are the
construction of knowledge-oriented information systems, improvement of research
organization processes, improvement of methods and tools for ontological analysis of
natural language objects using generative language models to extract knowledge from
them, applied aspects of using ontologies, meta-ontologies, knowledge integration systems
in transdisciplinary convergence clusters [6].
The proposed HCP-TM features a service-oriented architecture driven by ontologies [32,
33]. This architecture is implemented as a component-based collection of services, which
includes two foundational subsystems: the Medical-Rehabilitation Subsystem (MRS) and
the Information-Analytical Subsystem (IAS), also known as the Cognitive Subsystem.
The MRS includes essential functional modules such as the rehabilitation physician's
workstation, patient electronic cabinet, registration modules, and other necessary
components to support the rehabilitation process. A crucial part of the MRS is the
telemedical support subsystem for rehabilitation activities. The main tasks of remotely
controlled rehabilitation include establishing and refining optimal rehabilitation pathways,
forecasting and evaluating effectiveness based on the recovery of functions, and supporting
interaction among members of the interdisciplinary rehabilitation team.
The cognitive subsystem ensures the information-analytical processing of data
generated from spatially distributed network sources. This is achieved through continuous
interaction with all relevant specialists involved in the rehabilitation process using
interactive intelligent methods and tools implemented within the hybrid HCP-TM. The IAS,
a part of the patient-centric telerehabilitation cloud platform for oncology patients, is
known as MedRehabBot [6, 28, 34]. It is built on a specialized set of documents related to
physical and rehabilitation medicine and includes a suite of web services for context-
semantic analysis of textual documents, knowledge search and classification, ontology
generation in OWL, semantic trees, and graph-based knowledge bases within the domain.
MedRehabBot utilizes an information model based on a composite service represented
by a three-component tuple: web services and applications, information-technology
process service functions, and elements supporting the formation of an integrated
knowledge environment. The IAS, with its intelligent information-analytical support
functions, includes a comprehensive set of tools for evaluating the effectiveness and
improving rehabilitation strategies. This allows medical professionals, researchers, and
administrators to systematically enhance the telerehabilitation process, ensuring the
quality-of-service delivery and the best possible outcomes for patients undergoing
rehabilitation.
The information-analytical subsystem operates on the construction of a unified general
model, its precise mathematical substantiation, and solving a complex set of optimization
problems across the entire problem space.
From the perspective of project management, task distribution, and the functional use of
the architecture, the hybrid HCP-TM can be represented by three interacting subsystems:
the Medical-Rehabilitation Subsystem, the Information-Analytical Subsystem, and the
Telerehabilitation Subsystem.
The general functional architecture of the HCP-TM is shown in Figure 1.
The MRS subsystem includes key functional modules such as:
Physician’s Digital Workplace. Used by PM&R specialists, multidisciplinary
rehabilitation team members, specialized healthcare physicians, and primary care
physicians. This module provides a comprehensive digital interface for managing
patient care and coordinating with other healthcare professionals.
Patient’s Digital Cabinet. An online portal for patients to access their rehabilitation
plans, track progress, and communicate with their healthcare providers. This
module empowers patients by providing them with easy access to their
rehabilitation information and resources.
EHR Managing Module. Handles the storage, retrieval, and management of electronic
health records, ensuring secure and efficient access to patient data. This module is
critical for maintaining comprehensive and up-to-date medical records.
Administration and Registry Module. Manages the administrative tasks associated
with the telerehabilitation process, including scheduling, resource allocation, and
documentation. Maintains a registry of patients, treatments, outcomes, and other
relevant data (supports data collection and analysis, enabling the continuous
improvement of rehabilitation strategies). This module ensures the smooth
operation of the rehabilitation program by handling logistical and bureaucratic
aspects.
Figure 1: The General Functional Architecture of the HCP-TM.
The outlined configuration of the hybrid HCP-TM ensures the execution of several critical
functions: supporting the rehabilitation process (evaluating the patient's condition and
forming a rehabilitation diagnosis, predicting rehabilitation process indicators), building
the optimal rehabilitation route for the patient, developing methodological foundations for
supporting the rehabilitation process, and supporting interactive functions (doctor-patient-
system) in dialog mode. Additionally, it accumulates and intelligently processes information
from various sources, develops a set of modeling and optimization methods for the design
and application of the hybrid HCP-TM, analytically processes questionnaire data, and
statistically processes information.
The system also creates a "domains-scales" matrix, forms patient-centric sets of
procedures and supports them in real-time, and develops diagnostic gadgets for
determining the physiological and psychological state of patients. This includes video and
audio sessions for telerehabilitation. The sophisticated integration of these functions within
the HCP-TM framework provides a robust, patient-centered approach to telerehabilitation,
enabling continuous improvement and ensuring high-quality care.
4. The Three-layer Model of the Hybrid Cloud Platform for
Telerehabilitation Medicine
The Hybrid Cloud Platform for Telerehabilitation Medicine is built using an adapted three-
layer model of cloud service delivery and cloud computing [35]. This model comprises the
following layers.
Infrastructure Layer. Based on the Infrastructure as a Service (IaaS) model, the
Infrastructure Layer provides management capabilities for processing and storage
resources, communication networks, and other fundamental computing resources. This
layer supports the deployment and execution of various software, including operating
systems, application software, and system utilities. The infrastructure consists of three
main components:
Hardware. Includes servers, storage systems, client systems, and network
equipment.
Operating Systems and System Software. Comprises virtualization tools, automation
tools, and core resource management tools.
Middleware. Software for managing virtual operating systems.
Platform Layer. This layer, rooted in the Platform as a Service (PaaS) model, provides
access to information technology platforms. These platforms include operating systems,
database management systems, middleware, development, and testing tools hosted in the
cloud. The entire IT infrastructure, including computing networks and storage systems, is
managed by the provider. The provider determines the types of platforms available and the
set of manageable parameters. Developers can use these platforms to create virtual
instances, install, develop, test, and run application software while dynamically adjusting
the amount of consumed computing resources.
Service Layer. The Service Layer is based on the Software as a Service (SaaS) model. At
this layer, end-users (clients) access developed services and software via a thin client
(through a web browser) or an application programming interface (API). This access allows
users to leverage the functionalities of the platform without managing the underlying
infrastructure.
The overall architectural and structural organization diagram of the hybrid HCP-TM and
its components is shown in Figure 2. It includes the following components (hardware and
software, external services, interface, and network components):
HP ProLiant DL380p Gen8 Server – the high-performance server is a key component
of the Infrastructure layer of the HCP-TM. Located in a specialized room at the
Institute of Cybernetics, it ensures the reliability and high availability of the
platform’s services and resources. The server’s technical specifications include: CPU
– 2x Intel® Xeon® Processor E5-2695 v2, providing high performance and
multitasking capabilities; RAM – 400 GB Advanced ECC memory, enabling efficient
processing of large data volumes; Storage – 2x 400 GB SSDs in RAID 1 for additional
reliability, and 8x 400 GB SSDs in RAID 10 for optimized speed, and durability;
Network Connection – 1 Gbps, ensuring fast access to resources and data; Power
Supply – 2x 460 Watt power supplies, guaranteeing uninterrupted server operation;
Uninterruptible Power Supply – Eaton 5Cs 1500VA, protecting against power
outages and ensuring equipment operation during electrical failures.
Base OS – Ubuntu 22.04.3 LTS Jammy Jellyfish serves as the base OS for the server.
This version of Ubuntu is known for its stability, reliability, and wide range of
supported applications for workstations and servers. It is part of the Infrastructure
layer of the HCP-TM, ensuring the reliable operation and interaction of all platform
components.
Virtualization Module based on Kernel-based Virtual Machine (KVM) – KVM is a
high-performance virtualization solution integrated directly into the Linux kernel.
Designed specifically for the x86 architecture, KVM uses the capabilities of modern
Intel and AMD processors that support hardware virtualization through Intel VT
(Virtualization Technology) and AMD SVM (Secure Virtual Machine) technologies. A
key feature of KVM is its ability to run multiple virtual machines with different
operating systems on a single physical host, with each virtual machine using its own
Linux kernel, and resources being efficiently and flexibly allocated through
integration with the OS kernel. Infrastructure layer of the HCP-TM.
Virtual Environment Management Module LibVirt – this module, along with a set of
corresponding tools, provides unified management of virtual environments,
regardless of their location—locally or remotely. One of the key features of LibVirt
is its versatility and flexibility: it supports a wide range of virtualization systems,
including Xen, QEMU, KVM, LXC, Virtuozzo, Microsoft Hyper-V, and others. This
makes LibVirt an ideal choice for administrators and developers seeking a flexible
and efficient solution for managing virtualization in diverse environments.
Infrastructure layer of the HCP-TM.
Virtual OSs – various specialized operating systems are used at the Platform layer of
HCET. These systems play a key role in ensuring the stable and efficient operation of
all platform components, including services, modules, and subsystems. The list of
virtual OSs used at this level includes: Ubuntu 22.04.3 LTS Jammy Jellyfish; Alpine
Linux 3.18 (a lightweight and secure operating system ideal for containers – Docker,
Podman, Kubernetes); Microsoft Windows Server 2022 (a robust platform for
deploying enterprise applications); Microsoft Windows 10 Pro.
Proxy Server / VPN Server – An external virtual private server (VPS) that ensures
the functioning of a virtual private network (VPN) using the modern WireGuard
security protocol. WireGuard is distinguished by its high level of data protection and
optimized performance. Additionally, this server operates the Nginx Proxy Manager
[36] responsible for managing domains, SSL certificates, redirects, and streams. This
set of tools allows for reliable, secure, and flexible access to network resources, as
well as optimizing and automating web traffic management processes. Services layer
of the HCP-TM.
Domain Name Registrar – the NIC.UA service is responsible for the registration and
management of the cloud platform’s domain name – https://e-rehab.pp.ua.
Additionally, NIC.UA ensures the stability and security of subdomains used for
various services, modules, and platform components. Choosing this registrar
guarantees not only reliability but also ease of domain resource management, as
well as the ability to quickly expand and adapt to new requirements and user needs.
It should be noted that domain names in the pp.ua zone are provided free of charge
in Ukraine. Services layer of the HCP-TM.
Network Component of the HCP-TM – the network structure is based on the internal
network of the Institute of Cybernetics, characterized by a high level of isolation.
This closed network is integrated with the external Proxy Server/VPN Server
through a reliable VPN tunnel based on WireGuard technology. This configuration
allows for efficient and secure data exchange between the internal network of the
Institute of Cybernetics and the external internet, ensuring the confidentiality,
integrity, and availability of information.
The three-layer model of the HCP-TM, consisting of Infrastructure, Platform, and Service
layers, provides a robust framework for delivering telerehabilitation services. This model
leverages state-of-the-art technologies and methodologies to ensure scalability, reliability,
and high performance. The integration of specialized hardware, advanced operating
systems, virtualization technologies, and secure network configurations enables the HCP-
TM to meet the demanding requirements of telerehabilitation, particularly for oncology
patients. This comprehensive approach ensures that healthcare providers can deliver
effective, personalized, and efficient rehabilitation services, ultimately improving patient
outcomes and enhancing the quality of care.
Figure 2: The Comprehensive Architectural and Structural Diagram of the HCP-TM.
5. Conclusions and Further Research
This study examined modern technological approaches to telerehabilitation medicine and
its information and communication support. The application of a hybrid cloud platform with
an ontology-driven service-oriented architecture enables the creation of an effective
environment for remote interaction between medical personnel and patients. The
architectural and technological concept of the platform is based on the principles of an
advanced model of an automated research workstation. This concept has proven successful
in practical implementation, demonstrating its significant potential in the field of
telerehabilitation medicine.
The transition from traditional methods to the digitalization of scientific research opens
new opportunities for improving the quality of medical care and ensuring access to it
anytime and anywhere. The HCP-TM allows for a comprehensive integration of various
advanced technologies, including AI, IoMT, and multimodal chatbots, which collectively
enhance the telerehabilitation process. By facilitating seamless data management, real-time
patient monitoring, and interactive functionalities, the platform ensures that healthcare
providers can deliver personalized and effective care remotely.
Moreover, the use of ontology-driven frameworks within the HCP-TM enhances data
interoperability and supports sophisticated data analytics and decision-making processes.
This not only improves the accuracy of patient assessments and the effectiveness of
treatment plans but also contributes to the overall efficiency of healthcare delivery.
Further research in this direction will promote the development of telerehabilitation and
ensure its widespread application in medical practice. Continuous advancements in cloud
computing, IoMT, AI, and ontology-related technologies are expected to drive innovation in
telerehabilitation, making it more accessible and effective for patients globally. This
ongoing evolution will ultimately lead to better health outcomes and a higher quality of life
for individuals undergoing rehabilitation. Additionally, our future studies may encompass
the integration of hardware support into MedRehabBot dialogue system, using state of the
art circuitry type processor [38, 39]. and logical hardware technologies [40] for
implementation.
The practical success of the HCP-TM in integrating these technologies underscores its
potential to transform the landscape of telerehabilitation medicine. By leveraging the
strengths of a hybrid cloud architecture and advanced data management techniques, the
platform sets a new standard for remote healthcare services, paving the way for future
innovations and improvements in this critical area of medicine.
Acknowledgements
This study would not have been possible without the financial support of the National
Research Foundation of Ukraine (Open Funder Registry: 10.13039/100018227). Our work
was funded by Grant contract:
Development of the cloud-based platform for patient-centered telerehabilitation of
oncology patients with mathematical-related modeling [3], application ID: 2021.01/0136.
References
[1] O. Palagin, V. Velychko, K. Malakhov, O. Shchurov: Research and development
workstation environment: the new class of current research information systems. Proc.
of 11th International Conference UkrPROG 2018, Kyiv, Ukraine, May 20-24, 2018,
CEUR-WS.org, online CEUR-WS.org/Vol-2139/255-269.pdf.
[2] O. Palagin, K. Malakhov, V. Velychko, T. Semykopna, O. Shchurov: Hospital information
smart-system for hybrid e-rehabilitation. Proc. of 13th International Conference
UkrPROG 2022, Kyiv, Ukraine, October 11-12, 2022, CEUR-WS.org, online CEUR-
WS.org/Vol-3501/s50.pdf.
[3] O. Palagin, V. Kaverinskiy, A. Litvin, K. Malakhov, OntoChatGPT information system:
ontology-driven structured prompts for chatgpt meta-learning, Int. J. Comput. (2023)
170–183. doi:10.47839/ijc.22.2.3086.
[4] GitHub - malakhovks/ken: ken (ukrainian, english). URL:
https://github.com/malakhovks/ken.
[5] GitHub - malakhovks/docsim: UkrVectōrēs (former docsim). URL:
https://github.com/malakhovks/docsim.
[6] O. Palagin, V. Kaverinskiy, K. Malakhov, M. Petrenko, Fundamentals of the integrated
use of neural network and ontolinguistic paradigms: A comprehensive approach,
Cybern. Syst. Anal. (2024). doi:10.1007/s10559-024-00652-z.
[7] K. S. Malakhov, Letter to the editor – update from Ukraine: development of the cloud-
based platform for patient-centered telerehabilitation of oncology patients with
mathematical-related modeling, Int. J. Telerehabilitation 15.1 (2023).
doi:10.5195/ijt.2023.6562.
[8] Cancer in Ukraine, 2021-2022 Incidence, mortality, indicators of the oncology service.
Bulletin of the national cancer registry of Ukraine No. 24.
URL: http://www.ncru.inf.ua/publications/BULL_24/PDF/BULL_24.pdf.
[9] A. Mishra, N. Kumari, S. K. Bisoy, S. Sahoo, Integration of medical internet of things with
big data in healthcare industry, in: Connected e-Health, Springer International
Publishing, Cham, 2022, pp. 33–51. doi:10.1007/978-3-030-97929-4_2.
[10] A. A. El-Saleh, A. M. Sheikh, M. A. M. Albreem, M. S. Honnurvali, The Internet of Medical
Things (IoMT): opportunities and challenges, Wirel. Netw. (2024).
doi:10.1007/s11276-024-03764-8.
[11] J. Crowder, AI chatbots, Springer Nature Switzerland, Cham, 2024. doi:10.1007/978-3-
031-45509-4.
[12] A. Ghubaish, T. Salman, M. Zolanvari, D. Unal, A. K. Al-Ali, R. Jain, Recent advances in the
internet of medical things (iomt) systems security, IEEE Internet Things J. (2020) 1.
doi:10.1109/jiot.2020.3045653.
[13] M. P. Baroni, M. F. A. Jacob, W. R. Rios, J. V. Fandim, L. G. Fernandes, P. I. Chaves, I.
Fioratti, B. T. Saragiotto, The state of the art in telerehabilitation for musculoskeletal
conditions, Arch. Physiother. 13.1 (2023). doi:10.1186/s40945-022-00155-0.
[14] J. Pool, M. Indulska, S. Sadiq, Large language models and generative AI in telehealth: a
responsible use lens, J. Am. Med. Inform. Assoc. (2024). doi:10.1093/jamia/ocae035.
[15] A. M. Dakhel, A. Nikanjam, F. Khomh, M. C. Desmarais, H. Washizaki, An overview on
large language models, in: Generative AI for effective software development, Springer
Nature Switzerland, Cham, 2024, pp. 3–21. doi:10.1007/978-3-031-55642-5_1.
[16] M. Zhao, RoCo: Dialectic Multi-Robot Collaboration with LLM. URL: https://project-
roco.github.io/.
[17] A. A, F. Dahan, R. Alroobaea, W. Y. Alghamdi, Mustafa Khaja Mohammed, F. Hajjej,
Deema mohammed alsekait, K. Raahemifar, A smart IoMT based architecture for E-
healthcare patient monitoring system using artificial intelligence algorithms, Front.
Physiol. 14 (2023). doi:10.3389/fphys.2023.1125952.
[18] S. Ramesh, GitHub - medblocks/paco: for lighspeed gen AI hackathon. A guardian for
the patient. URL: https://github.com/medblocks/paco.
[19] T. Shaik, X. Tao, N. Higgins, L. Li, R. Gururajan, X. Zhou, U. R. Acharya, Remote patient
monitoring using artificial intelligence: Current state, applications, and challenges,
WIREs Data Min. Knowl. Discov. (2023). doi:10.1002/widm.1485.
[20] R. Uddin, I. Koo, Real-Time remote patient monitoring: A review of biosensors
integrated with multi-hop iot systems via cloud connectivity, Appl. Sci. 14.5 (2024)
1876. doi:10.3390/app14051876.
[21] A. V. Palagin, N. G. Petrenko, Methodological foundations for development, formation
and it-support of transdisciplinary research, J. Autom. Inf. Sci. 50.10 (2018) 1–17.
doi:10.1615/jautomatinfscien.v50.i10.10.
[22] O. Palagin, M. Petrenko, M. Boyko: Ontology-related complex for semantic processing
of scientific data. Proc. of 13th International Conference UkrPROG 2022, Kyiv, Ukraine,
October 11-12, 2022, CEUR-WS.org, online CEUR-WS.org/Vol-3501/s26.pdf.
[23] K. Malakhov, M. Petrenko, E. Cohn, Developing an ontology-based system for semantic
processing of scientific digital libraries, South Afr. Comput. J. 35.1 (2023).
doi:10.18489/sacj.v35i1.1219.
[24] M. Petrenko, E. Cohn, O. Shchurov, K. Malakhov, Ontology-Driven computer systems,
South Afr. Comput. J. 35.2 (2023). doi:10.18489/sacj.v35i2.17445.
[25] S. Geisler, C. Quix, A. Schmeink, D. Kensche, Ontology-based data integration: A case
study in clinical trials, in: Science, engineering, and biology informatics, WORLD
SCIENTIFIC, 2010, pp. 115–138. doi:10.1142/9789814307710_0006.
[26] O. Palagin, V. Kaverinsky, M. Petrenko, K. Malakhov, Digital health systems: ontology-
based universal dialog service for hybrid e-rehabilitation activities support, in: 2023
IEEE 12th international conference on intelligent data acquisition and advanced
computing systems: technology and applications (IDAACS), IEEE, 2023.
doi:10.1109/idaacs58523.2023.10348639.
[27] B. H. de Mello, S. J. Rigo, C. A. da Costa, R. da Rosa Righi, B. Donida, M. R. Bez, L. C.
Schunke, Semantic interoperability in health records standards: a systematic literature
review, Health Technol. 12.2 (2022) 255–272. doi:10.1007/s12553-022-00639-w.
[28] V. Kaverinsky, K. Malakhov, Natural language-driven dialogue systems for support in
physical medicine and rehabilitation, South Afr. Comput. J. 35.2 (2023).
doi:10.18489/sacj.v35i2.17444.
[29] A. Litvin, O. Palagin, V. Kaverinsky, K. Malakhov, Ontology-driven development of
dialogue systems, South Afr. Comput. J. 35.1 (2023). doi:10.18489/sacj.v35i1.1233.
[30] K. S. Malakhov, Innovative hybrid cloud solutions for physical medicine and
telerehabilitation research, Int. J. Telerehabilitation 16.1 (2024).
doi:10.5195/ijt.2024.6635.
[31] Technical requirements for hybrid cloud platform for telerehabilitation medicine,
2024. URL: https://cdn.e-rehab.pp.ua/u/Technical-requirements-hybrid-cloud-
platform.pdf.
[32] A. V. Palagin, Architecture of ontology-controlled computer systems, Cybern. Syst. Anal.
42.2 (2006) 254–264. doi:10.1007/s10559-006-0061-z.
[33] A. Litvin, V. Velychko, V. Kaverinsky: A new approach to automatic ontology generation
from the natural language texts with complex inflection structures in the dialogue
systems development. Proc. of 13th International Conference UkrPROG 2022, Kyiv,
Ukraine, October 11-12, 2022, CEUR-WS.org, online CEUR-WS.org/Vol-3501/s16.pdf.
[34] K. Malakhov, V. Kaverinskiy, GitHub - knowledge-ukraine/MedRehabBot. URL:
https://github.com/knowledge-ukraine/MedRehabBot.
[35] S. Bhowmik, Cloud computing, Cambridge University Press, 2017.
[36] GitHub - NginxProxyManager/nginx-proxy-manager (Docker container for managing
Nginx proxy hosts). URL: https://github.com/NginxProxyManager/nginx-proxy-
manager.
[37] K. S. Malakhov, Letter to the editor – update from Ukraine: development of the cloud-
based platform for patient-centered telerehabilitation of oncology patients with
mathematical-related modeling, Int. J. Telerehabilitation 15.1 (2023).
doi:10.5195/ijt.2023.6562.
[38] M. Petrenko, A. Kurgaev, Distinguishing features of design of a modern circuitry type
processor, Upr. Sist. I Mashiny 187.5 (2003), 16–19. URL:
https://www.scopus.com/record/display.uri?eid=2-s2.0-
0347622333&origin=inward&txGid=149869b0942d36862dd35d2c7fb39d
3a.
[39] M. Petrenko, A. Sofiyuk, Distinguishing features of design of a modern circuitry type
processor, Upr. Sist. I Mashiny 188.6 (2003) 48–57. URL:
https://www.scopus.com/inward/record.uri?eid=2-s2.0-
0442276898&partnerID=40&md5=44974b40409363e5fe4378e240149c5
2.
[40] S. L. Kryvyi, V. M. Opanasenko, S. B. Zavyalov, Algebraic operations on fuzzy sets and
relations in automata interpretation implemented by logical hardware, Cybern. Syst.
Anal. (2022). doi:10.1007/s10559-022-00497-4.