FAIR and GDPR Compliant Population Health Data Generation, Processing and Analytics

Abstract

Generating and analysing patient data in clinical settings is an inherently sensitive process, requiring collaborative effort between clinicians and informaticians to generate value from these data, while mitigating risks to the data subject. As a result, efforts in utilizing external patient data pose significant challenges. We propose a data-centric framework based on the FAIR principles and GDPR guidelines to enhance data management at the point of care. By using the process of data visiting, a cross-facility method for federated data analytics, we can automate generation of novel aggregate data which was previously not realizable. In two sequential studies we show that these techniques, supported by a data stewardship programme, increase community-wide involvement in data generation, improve transparency and trust, provide direct value and data ownership, and enable regulatory and ethically compliant, cross-national data visiting under curated accessibility patterns for federated analytics.

Full text

Copyright © 2022 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
FAIR and GDPR Compliant Population Health
Data Generation, Processing and Analytics
Ruduan Plug1, Yan Liang1, Mariam Basajja1, Aliya Aktau1, Putu Hadi
Purnama Jati2, Samson Yohannes Amare3, Getu Tadele Taye3, Mouhamad
Mpezamihigo4, Francisca Oladipo4, and Mirjam van Reisen1,2
1Leiden University, 2311 EZ Leiden, Netherlands
2Tilburg University, 5037 AB Tilburg, Netherlands
3Mekelle University, 231 Mekelle, Tigray, Ethiopia
4Kampala International University, 20000 Kampala, Uganda
r.b.f.plug@umail.leidenuniv.nl
Abstract. Generating and analysing patient data in clinical settings
is an inherently sensitive process, requiring collaborative effort between
clinicians and informaticians to generate value from these data, while
mitigating risks to the data subject. As a result, efforts in utilizing exter-
nal patient data pose significant challenges. We propose a data-centric
framework based on the FAIR principles and GDPR guidelines to en-
hance data management at the point of care. By using the process of
data visiting, a cross-facility method for federated data analytics, we
can automate generation of novel aggregate data which was previously
not realizable. In two sequential studies we show that these techniques,
supported by a data stewardship programme, increase community-wide
involvement in data generation, improve transparency and trust, pro-
vide direct value and data ownership, and enable regulatory and ethi-
cally compliant, cross-national data visiting under curated accessibility
patterns for federated analytics.
Keywords: FAIR Data ·GDPR ·Data Management ·Data Stewardship
·Clinical Data ·Biomedical Ontologies ·Data Federation ·Data Visiting
1 Introduction
The generation and management of clinical Electronic Health Record (EHR)
data requires strong safeguards on adherence to regulations, data security and
protection of patient privacy and confidentiality [19]. These factors complicate
facilitation of regional analytics and data exchange, which is seen as a criti-
cal factor in concerns of global and cross-national population health. Various
methods have been developed to address concerns of data security and privacy
protection [10, 32]. However, these methods tend to be problematic in practical
use and lack ontology-based standards for cross-facility interoperability and the
versatility to enable adherence to regulations set out by the relevant national
Ministry of Health (MoH) and regional legislature.

2 Plug et al.
The study implemented by the Virus Outbreak Data Network (VODAN)
Africa investigates the preparation and use of digital patient data in Africa. The
African continent is least represented in global health data, and the limitations
and challenges on digitisation of health data that lead to biases in globally
available data are well-documented [20]. Highly developed nations generate the
vast majority of the medical data, and as a result see the most representation
and benefit from research, while low-resource and rural areas tend to generate
few data and consequently are underrepresented in health research.
Efforts of developed nations to generate digital patient data from remote and
impoverished regions with vulnerable populations have often led to extractive
practices [15, 3], producing data sets that do not become available to or directly
serve the benefit of local populations and their health facilities. The transfer of
patient data aggregates from the facility where the data is produced to external
research facilities, poses ethical and legal concerns, in terms of the ownership of
the data and the link to the point of care [6].
These practices in data generation have led to a lack of trust due to the ab-
sence of standards in data ownership [14] and insufficiency of procedural trans-
parency within the generation and use of the data. Lack of capacity of data ana-
lytics within facilities compounds the problem of delaying adaptation of localized
information systems within clinics that can enable medical data generation and
regional clinical data exchange [1], while these localized data management prac-
tices at point of care are essential to the development of trustworthy and legally
compliant data generation methods [26, 23]. The lack of ownership and meaning-
ful use of the data further undermines the potential acceptance of digitisation of
patient data by the patient and other stakeholders. Hence, the quality and com-
pleteness of such data can be affected by the obstacles to adoptation of proposed
digitisation processes of electronic patient information [6].
Data management methods based on FAIR have been proposed as data lo-
calization strategies to improve the standards for patient data generation and
interoperability, and GDPR was utilized as a baseline standard to bridge the gap
to governance, which resulted in two studies implemented in Africa, spanning
from April 2020 - September 2020 and October 2020 to October 2021 [20].
2 FAIR and GDPR Standards
A central standard for regulatory frameworks within this study is encapsulated
by GDPR, which forms the basis of the initial trial by conceptualizing the point-
of-care as both data processor and data controller [4]. By using these standards,
explicit data ownership for the data subject and full control over data are pro-
vided at local levels while allowing for usage of these data under informed con-
sent. Within initial conversations with stakeholders across eight African nations,
including Tanzania, Uganda, Ethiopia, Somalia, Nigeria, Kenya, Tunisia and
Zimbabwe, this baseline was found to provide sufficient common ground while
being flexible to more stringent regulations layered upon GDPR as required by
local regulators [28].

FAIR and GDPR Compliant Health Data 3
Electronic
Patient Records
Medical
Measurements
Pseudo-
Anonymous
Data
Anonymous
Data
Processed
Data
Personal
Data
Data
Aggregated
Data
Accessibility
Specificity
Sensitivity
Fig. 1. Levels of Data Processing and Access Control.
An advantage of this approach is that GDPR in itself already provides a le-
gal framework to enable consent-based exchange of processed, anonymized data,
requiring an assessment of the relevance of the purpose of the data-collection.
However, to enable collaborative use of data such as federated analytics, we have
to look towards FAIR and ontology-based metadata to provide transparent, con-
sistent and machine-readable structure to data across different health facilities
[13], which can be sourced from HMIS already in use.
By ensuring FAIR compliance at the point of data generation, we provide
a set of transparent rules for permissions under which data can be found and
accessed, which is essential in forming trust in management of sensitive data.
A six-level system of access is illustrated in Figure 1, in which personal data
are not permissible to leave the facility while aggregated, processed and anony-
mous data may be exchanged through incremental levels of auditing required
before clearance is provided [9]. Interoperability is enabled through biomedical
ontologies, defined by research communities, providing the semantic links be-
tween data which can then be put into practice through metadata templating
[16]. The World Health Organization SMART guidelines recognize the relevance
of interoperable digital data use all of these levels, including the importance of
the meaningful use of data for quality health access at point of care [12].

4 Plug et al.
3 Study Results
To address concerns on security and privacy of patient data, which requires ca-
pacities to purposefully address the data production and assignment of responsi-
bilities regarding permission, a data stewardship programme was conceptualized
that aims to build a network of local experts on data management and gover-
nance [31]. Foundational to a versatile platform of trust and expertise in regard
to local and regional circumstances lies the interaction between human domain
experts and novel technology, and by bridging this gap, improvements in trust
and safety can be attained. Data stewards are primarily trained to handle data
management and auditing of data processing directly at the point of care.
Utilizing FAIR, assisted by biomedical ontology services such as NCBO Bio-
Portal [30], has already seen great potential in managing, analysing and reusing
biomedical samples across research facilities, for which we show an example in
Figure 2. Unique identifiers and data provenance support the documentation of
data ownership, while the use of common terminologies and semantics through
ontologies ensures that analytics across facilities is possible. Making such tech-
niques common practice for EHR data makes cross-facility and global analytics
of population health data possible without loss of data ownership or extensive
post-processing. This is critical for observational research with very limited data
such as rare diseases, which impose de-anonymization risks, or time-sensitive
analytics such as measuring incidence of COVID-19 across geographies.
The first study was conducted with universities within Africa, across Uganda,
Kenya, Ethiopia, Nigeria, Tunisia and Zimbabwe, in a collaboration of Kampala
International University (KIU), Tangaza University, Mekelle University, Addis
Ababa University, Ibrahim Badamasi University, University de Sousse, Great
Zimbabwe University (GZU), as well as the Leiden University Medical Center
(LUMC) [8, 20, 21] in Europe consisting of two core components. The first core
component was the sustainable data stewardship programme Training of Train-
ers (ToT) to train experts in data process curation and data management, based
on the FAIR principles under GO TRAIN [25].
The data stewards in turn are also equipped with skills to transfer this exper-
tise to other aspiring data experts, contributing to the UN sustainable develop-
ment goals [22]. The training program has resulted in 30 trained data stewards
whom can produce human and machine readable vocabulary relevant to patient
data records [28], from which ontologies can be defined that provide mappings
of data to semantics for FAIRification during point-of-care data production.
Adhering to the process of building expertise through ToT, the technological ar-
chitecture was developed and FAIR Data Point (FDP) services were established
within clinical settings at medical facilities. The FDPs were implemented using
local deployments of DS Wizard [17] to enable FAIR data production, for which
data generation was modelled on the WHO SARS-CoV-2 electronic Case Report
Forms (eCRF) ontology [2] stored as RDF graph databases.

FAIR and GDPR Compliant Health Data 5
Identifiers
Organism
Package
Attributes
BioProject
Submission
Pathogen:
clinical
or
host-associated
sample
from
Severe
acute
respiratory
syndrome
coronavirus
2
BioSample:
SAMN14656635;
Sample
name:
hCoV-19/USA/WI-179
/2020;
SRA:
SRS6514344
Severe
acute
respiratory
syndrome
coronavirus
2
Viruses;
Riboviria;
Orthornavirae;
Pisuviricota;
Pisoniviricetes;
Nidovirales;
Cornidovirineae;
Coronaviridae;
Orthocoronavirinae;
Betacoronavirus;
Sarbecovirus;
Severe
acute
respiratory syndrome-related
coronavirus
Pathogen: clinical or host-associated; version 1.0
strain hCoV-19/USA/WI-179/2020
isolate Homo sapien
collected by Milwaukee Public Health Department
collection date 2020-03-21
geographic location USA: Wisconsin, Milwaukee
host Homo sapiens
host disease COVID-19
isolation source nasal swab
latitude and longitude 43.042180 N 87.908670 W
ARTIC barcode identifiers NB03
PRJNA614504
Retrieve all samples from this project
UW-Madison, Shelby O'Connor; 2020-04-21
Accession: SAMN14656635 ID: 14656635
Fig. 2. An example of rich, ontology-assisted metadata and associated data already
being successfully applied and used to enable interoperability and reusability in
anonymized pathogen samples isolated from patients [24] (NCBI).
Following deployment, experiments were performed with local, in-residence
data production and subsequent cross-national SPARQL queries using the FAIR
data visiting model [18]. The first such clinical query utilizing the findability
and accessibility framework of FAIR was held on 29 September 2020 between
the FDPs at KIU and LUMC. This study demonstrated the feasibility of data-
querying of federated analytics across two continents, involving patient data held
in residence, curated and stored in the place where the data was produced.
A successful proof of concept was presented on international regulatory agree-
ments and a clinical implementation of the data ownership preserving framework
modelled using the FAIR concepts and GDPR. During this experiment, inter-
national cooperation and expertise was developed with focus on findability and
accessibility of clinical patient data, findable under well-specified and transparent
conditions. The aspects of interoperability and reusability were not operationally
implemented during this study and there was only one eCRF as an immutable
ontology which limited the flexibility of use.

6 Plug et al.
In direct continuation of the first trial, a second study was conducted to ad-
dress novel methods to combine ontology-assisted technology and community-
expertise in order to enable cross-facility interoperability and ultimately reusabil-
ity of data [20]. The second study period saw the number of participating nations
increase from six to eight including clinics and hospitals from Ethiopia, Kenya,
Nigeria, Somalia, Tanzania, Uganda, Tunisia, and Zimbabwe.
Essential to these efforts were retooling and deployment of localized CEDAR
[7] instances, which provide an open source platform assisted by BioPortal on-
tologies to produce, share and curate metadata templates and the data generated
from these templates in RDF format. This ensures that data has full providence
during production and provides interoperability through the open and transpar-
ent definitions of the ontologies. Different templates based on the same ontologies
are inherently interoperable on Common Data Elements (CDEs) [11], while data
from different ontologies can be matched by similarly utilizing common terms
and ontological semantic linkages [5, 27], which match the semantics from one
graph structure to another as a translation layer.
Data
Processor
Data
Controller
Regulatory
Data
Subject
Auditing
Data Steward
Production
Clinician
FAIRification Aggregation
Facility
Storage
Analytics
Fig. 3. FAIR and GDPR-based Framework for Data Processing and Analytics.
Central to the advantages offered by this approach are the engagement of
the scientific community, medical facilities, data stewards and legislature, which
all have been involved in the design and deployment of this architecture. In

FAIR and GDPR Compliant Health Data 7
addition, broad scale support was received from both the medical community
as well as the local MoHs [29]. During the second study, country coordinators
have been specified for each country to liaison with local facilities and MoHs,
while technical leads form the bridge between country coordinators and the
deployment. Data stewards are primarily tasked with guiding and auditing the
day-to-day operation of data generation and processing tasks.
The study pioneered a novel, fully FAIR and GDPR compliant, localized
health data generation procedure as a distributed network of FDPs that can
either function entirely independently or collaborate through data visiting pro-
cedures [20]. The resulting minimal viable product resolved the issue of data
ownership by fully FAIR local data production being conducted and utilizing
expertise from data stewards to conduct audits on data visiting requests, which
ensures that all data visiting queries, either to specific facilities or across all
indexed FDPs, comply to data ownership standards and regulations. This is fa-
cilitated by means of local data processing, such that the original data never
leaves the confinement of the medical facility, towards completely anonymized
processed data or aggregates modelled as federated analytics.
The complete procedure of this study is illustrated in Figure 3. This shows
the flow of data from the data owner, in this case the data subject, interpreted
by local clinicians, processed by data stewards using the FAIR data tooling
and then being made available in local storage. Often these data originate from
current health information systems such as DHIS2, from which data can also be
imported into CEDAR as JSON or RDF formatted data. Upon request for data
access using transparent accessibility procedures, under predefined conditions
and permission by the data controller, aggregated data can be made available
upon clearance of audit by the data steward.
4 Conclusion
During this study we have investigated, implemented and deployed a novel FAIR,
GDPR compliant data management architecture for curating, repositing and
analysing patient health data across health facilities. We have shown that by
using the FAIR principles, we can utilize biomedical ontologies to formally struc-
ture the data generation process through facility-catered metadata templating,
while retaining interoperability among data sources defined by these templates.
These formal specifications for interoperability provide an essential component
for privacy-oriented federated analytics across health facilities.
With this study we have identified the universal need for the recognition of
data ownership and control of patient data in relation to the health facilities
where data is produced, and the recognition of data origin and legal rights of
the patient as data subject. Data stewardship is proposed as a key instrument
in ensuring there is transparency, community-based trust and accountability for
repositing and processing patient data, as well as being instrumental to audit-
ing aggregated analytics performed on these data. This has shown encouraging
results with broad support from both health facilities and national MoHs.

8 Plug et al.
In addition, we recognize the importance of the locale of data generation.
By keeping full control over the data at the most localized level, we ensure that
data are handled in accordance with local regulations and ethical foundations.
Based on the support from legislature and research communities, we have found
evidence that doing so leads to a higher engagement in data production within
previously underserved communities. Broad engagement is essential in reducing
data bias and can encourage that aggregated data are being used and analysed
in a way that is meaningful within the local context.
By securely repositing data at the most localized level, while exposing cu-
rated, rich metadata under FAIR, we enable the possibility for federated data
analytics upon individual, controlled authorization without the risk of exposing
the underlying sensitive data. While generating FAIR data can be enabled using
a systematic ontology-matching approach, by linking the data generation process
to FAIR templates based on domain ontologies, the auditing of data processing
and analytical queries still requires significant knowledge and responsibility to
comply with ethical standards and local regulations, for which data stewardship
forms an essential area of local expertise.
Underlining these findings lies the importance between the relationship of
data generation and the in(direct) purpose of such data collection and processing
activities. Significant progress in EHR data analytics can be made by improving
the processes from the very origin of the data and ensuring that these processes
are transparent, well-defined and FAIR, which is in line with the SMART guide-
lines presented by the World Health Organization.
References
1. Basajja, M., Suchanek, M., Taye, G.T., Amare, S.Y., Mutwalibi, N., Folorunso, S.,
Plug, R., Oladipo, F., van Reisen, M.: Proof of concept and horizons on deployment
of fair in the covid-19 pandemic. Data Intelligence, Special Issue: Launching an
international FAIR data network for COVID data. Forthcoming. (2021)
2. Bonino, L.: Who covid-19 rapid version crf semantic data model, Retrieved
from: https://bioportal.bioontology.org/ontologies/COVIDCRFRAPID (accessed
9 September 2021)
3. Chu, K.M., Jayaraman, S., Kyamanywa, P., Ntakiyiruta, G.: Building research
capacity in africa: Equity and global health collaborations. PLoS Medicine 11
(2014)
4. Council, E.P..: Regulation (eu) 2016/679 on the protection of natural persons with
regard to the processing of personal data and on the free movement of such data,
and repealing directive 95/46/ec (general data protection regulation). Official Jour-
nal of the European Union 119, 1–88 (2016)
5. Euzenat, J., Shvaiko, P.: Ontology matching. In: Springer Berlin Heidelberg (2013)
6. Garrib, A., Stoops, N., Mckenzie, A., Dlamini, L.C., Govender, T., Rohde, J.E.,
Herbst, K.: An evaluation of the district health information system in rural south
africa. South African medical journal = Suid-Afrikaanse tydskrif vir geneeskunde
98 7, 549–52 (2008)
7. Gon¸calves, R.S., O’Connor, M., Romero, M.M., Egyedi, A.L., Willrett, D., Gray-
beal, J., Musen, M.: The cedar workbench: An ontology-assisted environment for

FAIR and GDPR Compliant Health Data 9
authoring metadata that describe scientific experiments. The semantic Web–ISWC:
International Semantic Web Conference proceedings. International Semantic Web
Conference 10588, 103–110 (2017)
8. Jacobsen, A., de Miranda Azevedo, R., Juty, N.S., Batista, D., Coles, S.J., Cornet,
R., Courtot, M., Crosas, M., Dumontier, M., Evelo, C.T.A., Goble, C.A., Guiz-
zardi, G., Hansen, K.K., Hasnain, A., Hettne, K.M., Heringa, J., Hooft, R.W.,
Imming, M., Jeffery, K.G., Kaliyaperumal, R., Kersloot, M.G., Kirkpatrick, C.R.,
Kuhn, T., Labastida, I., Magagna, B., McQuilton, P., Meyers, N., Montesanti, A.,
van Reisen, M., Rocca-Serra, P., Pergl, R., Sansone, S.A., da Silva Santos, L.O.B.,
Schneider, J., Strawn, G.O., Thompson, M., Waagmeester, A., Weigel, T., Wilkin-
son, M.D., Willighagen, E., Wittenburg, P., Roos, M., Mons, B., Schultes, E.: Fair
principles: Interpretations and implementation considerations. Data Intelligence 2,
10–29 (2020)
9. Jati, P.H.P., Flikkenschild, E., Meerman, B., Nodehi, S., Plug, R., Oladipo, F.,
van Reisen, M.: Data access, control, and privacy protection on vodan africa ar-
chitecture. Data Intelligence, Special Issue: Launching an international FAIR data
network for COVID data. Forthcoming. (2021)
10. Jin, H., Luo, Y., Li, P., Mathew, J.P.: A review of secure and privacy-preserving
medical data sharing. IEEE Access 7, 61656–61669 (2019)
11. Lin, C.H., Wu, N.Y., Liou, D.M.: A multi-technique approach to bridge electronic
case report form design and data standard adoption. Journal of biomedical infor-
matics 53, 49–57 (2015)
12. Mehl, G., Tun¸calp, ¨
O., Ratanaprayul, N., Tamrat, T., Barreix, M., Lowrance,
D.W., Bartolomeos, K., Say, L., Kostanjsek, N., Jakob, R., Grove, J.T., Mariano,
B., Swaminathan, S.: Who smart guidelines: optimising country-level use of guide-
line recommendations in the digital age. The Lancet. Digital health (2021)
13. Mons, B., Neylon, C., Velterop, J., Dumontier, M., da Silva Santos, L.O.B., Wilkin-
son, M.D.: Cloudy, increasingly fair; revisiting the fair data guiding principles for
the european open science cloud. Inf. Serv. Use 37, 49–56 (2017)
14. Musolino, N., Lazdi¸n, J., Toohey, J., IJsselmuiden, C.: Cohred fairness index for
international collaborative partnerships. The Lancet 385, 1293–1294 (2015)
15. Odekunle, F.F., Odekunle, R.O., Shankar, S.: Why sub-saharan africa lags in elec-
tronic health record adoption and possible strategies to increase its adoption in
this region. International Journal of Health Sciences 11, 59 – 64 (2017)
16. ran Park, J.: Metadata quality in digital repositories: A survey of the current state
of the art. Cataloging & Classification Quarterly 47, 213 – 228 (2009)
17. Pergl, R., Hooft, R.W., Such´anek, M., Knaisl, V., Slifka, J.: ”data stewardship
wizard”: A tool bringing together researchers, data stewards, and data experts
around data management planning. Data Sci. J. 18, 59 (2019)
18. Plug, R., Liang, Y., Aktau, A., Basajja, M., Oladipo, F., van Reisen, M.: Termi-
nology on a fair-framework for the virus outbreak data network. Data Intelligence,
Special Issue: Launching an international FAIR data network for COVID data.
Forthcoming. (2021)
19. Price, W.N., Cohen, I.G.: Privacy in the age of medical big data. Nature Medicine
25, 37–43 (2019)
20. van Reisen, M., Oladipo, F.O., Stokmans, M., Mpezamihgo, M., Folorunso, S.,
Schultes, E., Basajja, M., Aktau, A., Amare, S., Taye, G.T., Jati, P.H.P., Chindoza,
K., Wirtz, M., Ghardallou, M.E.L., van Stam, G., Ayele, W., Nalugala, R.M.,
Abdullahi, I., Osigwe, O., Graybeal, J., Medhanyie, A.A., Kawu, A.A., Liu, F.,
Wolstencroft, K., Flikkenschild, E., Lin, Y., Stocker, J.W., Musen, M.A.: Design

10 Plug et al.
of a fair digital data health infrastructure in africa for covid-19 reporting and
research. Advanced Genetics (Hoboken, N.j.) 2(2021)
21. van Reisen, M., Stokmans, M., Basajja, M., Ong’ayo, A., Kirkpatrick, C.R., Mons,
B.: Towards the tipping point for fair implementation. Data Intelligence 2, 264–275
(2020)
22. van Reisen, M., Stokmans, M., Mawere, M., Basajja, M., Ong’ayo, A., Nakazibwe,
P., Kirkpatrick, C.R., Chindoza, K.: Fair practices in africa. Data Intelligence 2,
246–256 (2020)
23. Sahay, S., Rashidian, A., Doctor, H.V.: Challenges and opportunities of using dhis2
to strengthen health information systems in the eastern mediterranean region: A
regional approach. The Electronic Journal of Information Systems in Developing
Countries 86 (2020)
24. Schriml, L.M., Chuvochina, M., Davies, N., Eloe-Fadrosh, E.A., Finn, R.D., Hugen-
holtz, P.B., Hunter, C.I., Hurwitz, B.L., Kyrpides, N.C., Meyer, F., Mizrachi, I.K.,
Sansone, S.A., Sutton, G.G., Tighe, S.W., Walls, R.B.: Covid-19 pandemic reveals
the peril of ignoring metadata standards. Scientific Data 7(2020)
25. Schultes, E., Mons, A., Mons, B., Kuzak, M., van Gelder, C., Hodson, S., Meer-
man, B., Jansen, M., Bonino, L.O., Drefs, I., Kriegel, K., Sustkova, H.P., Dumon-
tier, M., de Miranda Azevedo, R., Presutti, V.: Go train pillar, Retrieved from:
https://osf.io/za96b/ (accessed 8 September 2021)
26. Shaffer, J.G., Doumbia, S., Ndiaye, D., Diarra, A., Gomis, J.F., Nwakanma, D.C.,
Abubakar, I., Ahmad, A., Affara, M., Lukowski, M., Valim, C., Welty, J.C., Mather,
F.J., Keating, J., Krogstad, D.J.: Development of a data collection and manage-
ment system in west africa: challenges and sustainability. Infectious Diseases of
Poverty 7(2018)
27. Shvaiko, P., Euzenat, J.: Ontology matching: State of the art and future challenges.
IEEE Transactions on Knowledge and Data Engineering 25, 158–176 (2013)
28. VODAN-Africa: About vodan africa, Retrieved from: https://www.vodan-
totafrica.info/vodan-africa.php?i=1&a=about-vodan-africa (accessed 7 September
2021)
29. VODAN-Africa: Vodan letters of support, Retrieved from: https://www.vodan-
totafrica.info/vodan-africa.php?i=15&a=letters-of-support (accessed 29 Septem-
ber 2021)
30. Whetzel, P.L., Noy, N., Shah, N.H., Alexander, P.R., Nyulas, C., Tudorache, T.,
Musen, M.A.: Bioportal: enhanced functionality via new web services from the
national center for biomedical ontology to access and use ontologies in software
applications. Nucleic Acids Research 39, W541 – W545 (2011)
31. Wilkinson, M.D., Dumontier, M., Aalbersberg, I.J., Appleton, G., Axton, M.,
Baak, A., Blomberg, N., Boiten, J.W., da Silva Santos, L.O.B., Bourne, P.E.,
Bouwman, J., Brookes, A.J., Clark, T., Crosas, M., Dillo, I., Dumon, O., Ed-
munds, S.C., Evelo, C.T.A., Finkers, R., Gonz´alez-Beltr´an, A.N., Gray, A.J.G.,
Groth, P., Goble, C.A., Grethe, J.S., Heringa, J., ‘t Hoen, P.A.C., Hooft, R.W.W.,
Kuhn, T., Kok, R.G., Kok, J.N., Lusher, S.J., Martone, M.E., Mons, A., Packer,
A.L., Persson, B., Rocca-Serra, P., Roos, M., van Schaik, R.C., Sansone, S.A.,
Schultes, E., Sengstag, T., Slater, T., Strawn, G.O., Swertz, M.A., Thompson, M.,
van der Lei, J., van Mulligen, E.M., Velterop, J., Waagmeester, A., Wittenburg,
P., Wolstencroft, K., Zhao, J., Mons, B.: The fair guiding principles for scientific
data management and stewardship. Scientific Data 3(2016)
32. Yang, J.J., Li, J., Niu, Y.: A hybrid solution for privacy preserving medical data
sharing in the cloud environment. Future Gener. Comput. Syst. 43-44, 74–86
(2015)