International Journal of Engineering and Advanced Technology Studies 13 (1), 22-28, 2025
Print ISSN: 2053-5783(Print)
Online ISSN: 2053-5791(online)
Website: [Link]
Publication of the European Centre for Research Training and Development -UK
Interoperability within Healthcare Systems through
FHIR, Artificial Intelligence and Cloud Integration
Malvina Halilaj, PhD
Informatics Department, Faculty of Natural Sciences, Tirana University
Dr. Suela Vasil (Maxhelaku)
Informatics Department, Faculty of Natural Sciences, Tirana University
Anxhela Kosta, PhD
Informatics Department, Faculty of Natural Sciences, Tirana University
Ilma Lili, PhD
Informatics Department, Faculty of Natural Sciences, Tirana University
doi: [Link] Published April 16, 2025
Citation: Halilaj M., Vasil S., Kosta A. and Lili I. (2025) Interoperability within Healthcare Systems through FHIR,
Artificial Intelligence and Cloud Integration, International Journal of Engineering and Advanced Technology Studies, 13
(1), 22-28
Abstract: One of the main challenges for healthcare interoperability is defining common standards for the
structured content of healthcare data and the transport of this data between different systems. The purpose of
this paper is to enable the exchange of these data, with a focus on the FHIR (Fast Healthcare Interoperability
Resources) protocol which created the HL7 (Health Level Seven International) standard, a framework that has
become widely adopted by all stakeholders around the world to determine the content of this data and the way
they want to share it and to identify the potential of the integration of Artificial Intelligence with the HL7 FHIR
standart. The integration of this data was one of the main questions raised in the project “Integration of AI in
Advancing Interdisciplinary Research at the University of Tirana”.
Keywords: FHIR, HL7, interoperability, data integration, artificial intelligence, interdisciplinary research
INTRODUCTION
The health sector generates a large volume of data from patient files and scans to genetic data and wearable
devices. Fast Healthcare Interoperability Resources (FHIR) has been emerged as the standard that is leading for
healthcare data exchange, providing a framework that is common for structuring and for accessing this data. At
the same time, artificial intelligence (AI) techniques are transforming how we analyse as well as derive
understandings from healthcare information. The integration of these domains represents a powerful tool
The exchange of clinical records electronically is one of the most import aspects and challenge in the healthcare
sector. The integration is important not only within a healthcare centre, but also between different healthcare
centres, along with for patients in order to easily access their own medical data. Standardized agreements, along
with a set of common rules, are necessary to ease this exchange. Efficient, accurate and secure data exchange,
along with the preservation of its integrity, confidentiality, as well as availability, results from these standards.
Faster and more reliable system interactions are achieved through adherence to these frameworks.
22
International Journal of Engineering and Advanced Technology Studies 13 (1), 22-28, 2025
Print ISSN: 2053-5783(Print)
Online ISSN: 2053-5791(online)
Website: [Link]
Publication of the European Centre for Research Training and Development -UK
To address these important needs, is being used HL7 (Health Level Seven), an important set of international
standards for the transfer of clinical and administrative data between software applications used by different
healthcare providers. Many hospitals, along with other healthcare organizations, use quite a few, very diverse
computer systems for a truly wide range of purposes, from billing records to patient tracking. Some of these
systems fail to communicate (or "interface") upon receiving or requesting new information. HL7 thus needed a
new method to easily exchange clinical data, which they called FHIR. It is modern web-based API technology
that simplifies implementation using a HTTP-based RESTful protocol plus HTML along with CSS for user
interface integration plus a choice of JSON XML or RDF for data representation as well as Atom for results.
This eases interoperability between legacy healthcare systems, enabling easy healthcare information access for
providers and people across multiple devices and allowing third-party developers to easily integrate medical
applications into existing systems (Mohammad [Link], 2020). The "Integration of Artificial Intelligence in
advancing interdisciplinary research at the University of Tirana" project promotes cross-disciplinary
collaboration and innovation in AI applications. The project also aims to raise awareness and understanding of
AI and implement AI in various disciplines. One of the key areas of the project was healthcare. This study is
focused improving interoperability within healthcare systems by using artificial intelligence.
LITERATURE REVIEW
Many healthcare organizations use HL7 FHIR (Fast Healthcare Interoperability Resources), a standard dealing
with many interoperability challenges in healthcare information exchange (Maxhelaku & Kika, 2019; Sharma
& Aggarwal, 2018). Open-source tools, such as the HAPI FHIR server (Hussain et al., 2018), allow for the
straightforward implementation of FHIR, which readily employs RESTful principles. Evidence from various
healthcare sectors (Jaffe et al., 2023) demonstrates the worldwide implementation of this standard. Various
contexts acknowledge FHIR's capacity to enhance interoperability. Modernization initiatives are significantly
in progress to revise antiquated standards like HL7 v2.x and ensure their compatibility with the continuously
advancing FHIR (Oemig, 2019).
The integration of AI and HL7 FHIR significantly enhances the interoperability of healthcare data and supports
better decision-making processes. Saripalle, also noted the standard developed by HL7 which facilitates
seamless data exchange across healthcare systems. By using FHIR data, AI analytical models can improve the
predictive abilities and streamline healthcare outcomes (Lekkala, 2023). Interoperable applications, even in
developing nations using paper-based systems (Ahmad et al., 2018), are eased by the SMART on FHIR
platform. Multiple aspects of healthcare, such as mortality reporting (Hoffman et al., 2018; Hoffman et al.,
2017), can be importantly improved by the use of these applications. The creation of incredibly useful semantic
interoperability platforms, specifically designed for AI-driven applications using HL7 FHIR (Rigas et al., 2024),
has importantly advanced the field, along with the implementation of strong smart contract authentication via
the graph-based FHIR architecture, further improving the efficiency and security of e-healthcare systems
(Sreejith & Senthil, 2023). These improvements are driving important improvements in healthcare, increasing
efficiency, encouraging interoperability, as well as promoting data-centric approaches.
Cloud platforms support FHIR implementation into improving healthcare data interoperability and analytics.
Serverless FHIR machine learning model deployment architectures have been proposed (B. Eapen et al., 2020)
and integrating FHIR with IoT cloud platforms for personal health records (Jae-Ki Hong et al., 2017). FHIR
allows cloud-based solutions such as HealtheDataLab to show how advanced analytics can be used with large-
scale healthcare data (L. Ehwerhemuepha et al., 2020). In addition, frameworks for implementing SMART on
FHIR in developing countries have been developed by researchers (Abrar Ahmad et al. 2018). FHIR's
application in several clinical and research contexts has also been explored by many researchers (Ashley C.
Griffin et al. 2022). Researchers have importantly addressed security concerns, incorporating role-based access
control into FHIR (Yaira K. Rivera Sánchez et al., 2017), as well as proposing highly scalable cloud
23
International Journal of Engineering and Advanced Technology Studies 13 (1), 22-28, 2025
Print ISSN: 2053-5783(Print)
Online ISSN: 2053-5791(online)
Website: [Link]
Publication of the European Centre for Research Training and Development -UK
architectures for EHR interoperability (Derlis Gómez et al., 2021). FHIR's potential to improve healthcare data
management and analysis in cloud environments is highlighted by these improvements.
Implementation of the FHIR on cloud
In the project "Integration of AI in Advancing Interdisciplinary Research at the University of Tirana", we have
used FHIR standard for health data exchange by focusing on the data format being exchanged, its storage, and
interoperability. The cloud has seen significant advancement in the implementation of FHIR. The Azure API
for FHIR enables users to facilitate data exchange through an FHIR API, while use a Platform as a Service
(PaaS) solution in Azure, specifically tailored for the management and storage of PHI data in its native FHIR
format. Through this provision, Microsoft will manage the operations, maintenance, updates, and compliance
requirements. This allows developers to concentrate on creating solutions with health data while utilizing other
platform services offered on Azure, including Data Factory, Storage, and Data Bricks for analytical purposes.
Utilizing data in FHIR format, the FHIR Server for Azure allows developers to efficiently query and oversee
FHIR data in the cloud, monitor and control data access, and standardize data for machine learning applications.
(Bender, Sartipi, 2013).
The architecture presented below aims to demonstrate the process of linking a web application with the FHIR
API. Additionally, functionalities like SMART are included in the FHIR Active Directory Proxy. The
deployment can be accomplished through the Azure API for the FHIR PaaS server.
Figure [Link] Architecture on Azure.
With all necessary resources established in Azure, it is possible to proceed to generate patient data. To obtain
the most accurate data, we have used SyntheaTM, a Symptomatic Simulator designed for the Patient Population.
The objective is to obtain synthetic, realistic (yet not actual) patient data along with associated health data
presented in multiple formats. JSON is the format of the data used. The following presents an overview of a
patient's information, their consultation with the doctor, and the patient's health status.
24
International Journal of Engineering and Advanced Technology Studies 13 (1), 22-28, 2025
Print ISSN: 2053-5783(Print)
Online ISSN: 2053-5791(online)
Website: [Link]
Publication of the European Centre for Research Training and Development -UK
resource": {"resourceType": "Patient",
"id": "318504cc-bfe7-4793-a221-83503d725476",
"value": "36704efd-6b33-4827-935f-8d1a01e576f1" ],
"resource": {
"name": [ "resourceType": "Encounter",
{ "use": "official", "id": "b6c84e45-163c-4912-9afb-3cdd127d7f24",
"status": "finished",
"family": "Halilaj",
"given": ["Malvina"], "coding": [
"prefix": ["Ms."] {
"system": "[Link]
} "code": "162673000",
], "display": "General examination of patient (procedure)"
}
"telecom": [ ],
{"system": "phone", "text": "General examination of patient (procedure)"
"value": "555-521-5778", }
],
"use": "home"}], "subject": {
"gender": "female", "reference": "urn:uuid:2ea47dc8-0c8e-4156-9c7c-a0968636fe11",
"display": "[Link] Halilaj"
"birthDate": "1996-04-11", },
"address": [ "participant": [
{"line": [ {
"individual": {
"420 Ryan Trailer Unit 45" "reference": "urn:uuid:0000016d-4b00-3a51-0000-0000000157fc",
], "display": "Dr. Kamala553 Schinner682"
}
"city": "Tirane", }
"state": "Albania", ],
"postalCode": "1001", "period": {
"start": "2019-07-13T11:37:28+01:00",
"country": "AL" "end": "2019-07-13T11:52:28+01:00"
} },
"serviceProvider": {
], "reference": "urn:uuid:7b76c6b3-c413-3293-914a-fcf8195f32bc",
"request": { "display": "HALLMARK HEALTH MEDICAL ASSOCIATES INC"
"method": "POST", }
},
"url": "Patient" "request": {
} "method": "POST",
"url": "Encounter"
}, }
},
Figure 2. Patient information and consultation details
"resourceType": "Condition",
"id": "f4984d60-0527-4a69-921a-170605c75464",
"clinicalStatus": {
"coding": [
{
"system": "[Link]
clinical",
"code": "active"
}
]
},
"code": {
"coding": [
{
"system": "[Link]
"code": "59621000",
"display": "Hypertension"
}
],
"text": "Hypertension"
},
"request": {
"method": "POST",
"url": "Condition"
Figure 3. Health conditions details.
25
International Journal of Engineering and Advanced Technology Studies 13 (1), 22-28, 2025
Print ISSN: 2053-5783(Print)
Online ISSN: 2053-5791(online)
Website: [Link]
Publication of the European Centre for Research Training and Development -UK
Integration of AI with FHIR
Integration of AI algorithms with the FHIR standards can improve interoperability, improve clinical decision
support, and increase outcome prediction for the patient. Recent studies shows that FHIR‐based analytics
framework increased data interoperability from 11% to 66%. (Ayaz [Link], 2023). A multi‐center development
showed that because of a deep learning model using the FHIR-formatted EHR data achieved AUROC values of
0.93–0.95 for in-hospital mortality, 0.75–0.77 for 30-day readmission, and 0.85–0.86 for prolonged length of
stay, performance metrics that exceed those from customary predictive models. (Rajkomar, et. al, 2023). These
findings support that AI-FHIR systems, via certain integration plans, can improve data standardization and
predictive accuracy compared to standard data integration [Link] were several articles that has been
reviewed based on the implementation of the AI algorithms (machine learning, deep learning or natural language
processing) integrated with FHIR standards. Some of the characteristics of the included studies are as follows:
Study AI-FHIR Integration approach
Ayaz et al., FHIR-based data Mapping algorithm to transform FHIR
2023 analytics framework resource data into relational database format
Balch et al., FHIR-based data Conversion of disparate data sources to FHIR
2023 analytics framework standards, use of FHIR Application
Programming Interfaces (APIs)
Rajkomar et FHIR-based data Representation of entire EHR in temporal
al., 2018 analytics framework order using FHIR standard
Rigas et al., AI-based applications Adoption of common semantic data models
2024 platform using Health using HL7 FHIR standard
Level Seven (HL7)
FHIR
Semenov et FHIR-based clinical Use of a FHIR Adapter to convert data from
al., 2019 decision support system FHIR format into an internal format
Tarumi et AI-driven clinical Substitutable Medical Applications, Reusable
al., 2021 decision support for Technologies (SMART) on FHIR framework,
diabetes using FHIR HAPI-FHIR API, OpenCDS framework
Williams et FHIR-based data Python-based mapping of hospital data to
al., 2022 harmonization pipeline FHIR concepts, saving in JavaScript Object
for AI applications Notation (JSON) format
RESULTS
Significant advancements in interoperability, security, and scalability have been realized with the adoption of
the FHIR standard in the exchange of healthcare data. This research demonstrated seamless healthcare data
interchange using FHIR and standardized formats such as JSON. The integration of various formats into
multiple cloud-based platforms has led to effective data transfer while preserving accuracy and integrity.
This paper was focused on the management of patient health information through the successful integration of
the FHIR server with the Azure API for FHIR. The creation of synthetic patient data was demonstrated by
using SyntheaTM, a simulator for patient populations. This simulation produced exceptionally realistic, entirely
26
International Journal of Engineering and Advanced Technology Studies 13 (1), 22-28, 2025
Print ISSN: 2053-5783(Print)
Online ISSN: 2053-5791(online)
Website: [Link]
Publication of the European Centre for Research Training and Development -UK
non-sensitive patient data that closely reflects the clinical workflows. The data were organized in FHIR format,
facilitating its application for testing various interoperability scenarios and for querying numerous patient
conditions, medical visits, and consultations.
CONCLUSION AND FUTURE RESEARCH
AI-enabled FHIR integration offers a huge promise in terms of making the healthcare systems smarter as well
as more responsive and patient-centered. AI-enabled FHIR integration shows a mixture of standardized
healthcare data models with detailed analytical techniques. Healthcare providers are able to deliver better
outcomes by bridging the gap between exchanging the data and intelligent analysis. They can optimize resources
and even reduce burden for the clinician. The combination creates new opportunities to better patient care,
optimize clinical work, and accelerate the research in medicine.
As FHIR keeps developing with several resources as well as implementation patterns, the opportunities for AI
integration will only expand. The key challenges remain ensuring the quality of the data, addressing privacy
concerns, integrating predictions into clinical workflows.
Challenges linked to healthcare data interoperability are importantly addressed via the major role of the FHIR
standard. FHIR bridges any gap between platforms and modern technologies by easing structured, secure, as
well as standardized data exchange across diverse systems.
Acknowledgments:
Our sincere gratitude to the generous support of National Agency for Scientific Research and Innovation
(NASRI) in Albania, which enabled us to carry out this study. The financial support received for our project "
Integration of AI in Advancing Interdisciplinary Research at the University of Tirana", based on Decision No.
10, date 21.07.2023, "On the approval of the financing of winning projects of the National Research and
Development Program for the Period 2023-2024", is responsible for the significant success of the study.
REFERENCES
Ayaz M, Pasha MF, Alahmadi TJ, Abdullah NNB, Alkahtani HK. Transforming Healthcare Analytics with
FHIR: A Framework for Standardizing and Analyzing Clinical Data. Healthcare. 2023; 11(12):1729.
[Link]
Ahmad, A., Azam, F., & Anwar, M.W. (2018). Implementation of SMART on FHIR in Developing Countries
Through SFPBRF. International Conferences on Biomedical and Bioinformatics Engineering.
Balch JA, Ruppert MM, Loftus TJ, Guan Z, Ren Y, Upchurch GR, Ozrazgat-Baslanti T, Rashidi P, Bihorac A,
Machine Learning–Enabled Clinical Information Systems Using Fast Healthcare Interoperability
Resources Data Standards: Scoping Review JMIR Med Inform 2023;11:e48297, doi: 10.2196/48297,
PMID: 37646309, PMCID: 10468818.
Bender ,D. and Sartipi,K. (2013) HL7 FHIR: an agile and restful approach to healthcare information exchange.
Eapen, B.R., Sartipi, K., & Archer, N.P. (2020). Serverless on FHIR: Deploying machine learning models for
healthcare on the cloud. ArXiv, abs/2006.04748.
Ehwerhemuepha, L., Gasperino, G., Bischoff, N., Taraman, S., Chang, A., & Feaster, W.W. (2020).
HealtheDataLab – a cloud computing solution for data science and advanced analytics in healthcare
with application to predicting multi-center pediatric readmissions. BMC Medical Informatics and
Decision Making, 20.
27
International Journal of Engineering and Advanced Technology Studies 13 (1), 22-28, 2025
Print ISSN: 2053-5783(Print)
Online ISSN: 2053-5791(online)
Website: [Link]
Publication of the European Centre for Research Training and Development -UK
Gómez, D., Romero, J., López, P., Vázquez, J., Cappo, C., Pinto, D., & Villalba, C. (2021). Cloud architecture
for electronic health record systems interoperability. Technology and health care: official journal of the
European Society for Engineering and Medicine.
Griffin, A.C., He, L., Sunjaya, A.P., King, A.J., Khan, Z., Nwadiugwu, M.C., Douthit, B.J., Subbian, V.,
Nguyen, V., Braunstein, M.L., Jaffe, C., & Schleyer, T. (2022). Clinical, technical, and implementation
characteristics of real-world health applications using FHIR. JAMIA Open, 5.
Hoffman, R.A., Wu, H., Venugopalan, J., Braun, P., & Wang, M.D. (2018). Intelligent Mortality Reporting
With FHIR. IEEE Journal of Biomedical and Health Informatics, 22, 1583-1588.
Hoffman, R.A., Wu, H., Venugopalan, J., Braun, P., & Wang, M.D. (2017). ‘Intelligent mortality reporting
with FHIR. 2017 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI),
181-184.
Hong, J., Morris, P., & Seo, J. (2017). Interconnected Personal Health Record Ecosystem Using IoT Cloud
Platform and HL7 FHIR. 2017 IEEE International Conference on Healthcare Informatics (ICHI), 362-
367.
Hussain, M., Langer, S.G., & Kohli, M.D. (2018). Learning HL7 FHIR Using the HAPI FHIR Server and Its
Use in Medical Imaging with the SIIM Dataset. Journal of Digital Imaging, 31, 334 - 340. ‘
Jaffe, C., Vreeman, D.J., Kaminker, D., & Nguyen, V. (2023). Implementing HL7 FHIR. Journal of Healthcare
Management Standards.
Maxhelaku, S., & Kika, A. (2019). Improving interoperability in healthcare using hl7 FHIR. Proceedings of
the 47th International Academic Conference, Prague.
Mohammad Jafari,Kathleen Connor, John M. Davis, Christopher Shawn (2020). Sub-Resource Access in FHIR
Lekkala, L.R. (2023). How AI Analytical Models Can Use FHIR (Fast Healthcare Interoperability Resources)
Data. Voice of the Publisher.
Oemig, F. (2019). HL7 Version 2.x Goes FHIR. Studies in health technology and informatics, 267, 93-98.
Rajkomar, A., Oren, E., Chen, K. et al. Scalable and accurate deep learning with electronic health records. npj
Digital Med 1, 18 (2018). [Link]
Rigas, E.S., Lagakis, P., Karadimas, M., Logaras, E., Latsou, D., Hatzikou, M., Poulakidas, A., Billis, A., &
Bamidis, P.D. (2024). Semantic interoperability for an AI-based applications platform for smart
hospitals using HL7 FHIR. J. Syst. Softw., 215, 112093.
Sánchez, Y.K., Demurjian, S.A., & Baihan, M.S. (2017). Achieving RBAC on RESTful APIs for Mobile Apps
Using FHIR. 2017 5th IEEE International Conference on Mobile Cloud Computing, Services, and
Engineering (MobileCloud), 139-144.
Saripalle, R. (2019). Fast Health Interoperability Resources (FHIR): Current Status in the Healthcare
System. Int. J. E Health Medical Commun., 10, 76-93.
Semenov, I., Osenev, R., Gerasimov, S., Kopanitsa, G., Denisov, D., & Andreychuk, Y. (2020). Experience in
Developing an FHIR Medical Data Management Platform to Provide Clinical Decision
Support. International Journal of Environmental Research and Public Health, 17(1), 73.
[Link]
Sharma, M., & Aggarwal, H. (2018). HL-7 Based Middleware Standard for Healthcare Information System:
FHIR. Proceedings of 2nd International Conference on Communication, Computing and Networking.
Sreejith, R., & Senthil, S. (2023). Smart Contract Authentication assisted GraphMap-Based HL7 FHIR
architecture for interoperable e-healthcare system. Heliyon, 9 4, e15180.
Tarumi Sh. et. al (2021). Leveraging Artificial Intelligence to Improve Chronic Disease Care: Methods and
Application to Pharmacotherapy Decision Support for Type-2 Diabetes Mellitus. Methods of
Information in Medicine
Williams, E. et, a. (2022). FHIR-DHP: A Standardized Clinical Data Harmonisation Pipeline for scalable AI
application deployment doi: [Link]
28