RadLexUA: a grassroots localization of the RSNA radiology lexicon for the Ukrainian language — technical foundations, lexical variability, and the institutional gap
DOI:
https://doi.org/10.66636/gmj.v1.i2.a115Keywords:
RadLex, radiology ontology, Ukrainian language, structured reporting, clinical NLP, eHealth UkraineAbstract
Background Radiological reporting in Ukraine remains predominantly free-text, with no nationally mandated structure or obligatory terminology. No Slavic language currently has a complete localization of RadLex, the Radiological Society of North America (RSNA) reference ontology of radiology terminology. The absence of a standardized Ukrainian vocabulary impedes structured reporting, cross-institutional data retrieval, clinical natural language processing, and alignment with emerging European Health Data Space interoperability requirements.
Methods RadLexUA was constructed through a systematic translation pipeline producing Ukrainian preferred names for 46,846 RadLex concepts and translated definitions for 3,840 of them, validated against a corpus of approximately 49,000 anonymized computed tomography, magnetic resonance imaging, ultrasound, and X-ray reports. A ten-stage natural language processing pipeline was developed encompassing anonymization, lemmatization, dictionary-based named entity recognition, fuzzy and semantic matching, and human-in-the-loop validation, anchoring annotations to RadLex Identifiers (RIDs).
Results The resource produced Ukrainian preferred names for 99.9% of RadLex concepts (46,801 of 46,846 automatically translated, 45 flagged for manual rendering). Corpus validation documented substantial lexical variability, with multiple clinically near-equivalent Ukrainian surface forms mapping to single RIDs. Eleven entries (0.29% of definitions) have received formal human sign-off by a reviewer with a radiology background; the remainder are at beta quality. The pipeline emits HL7 FHIR Observation resources keyed on RIDs, enabling downstream structured reporting and cohort retrieval.
Conclusion RadLexUA demonstrates that a usable technical substrate for Ukrainian radiology terminology standardization can be assembled without institutional infrastructure. However, the distance between a translated ontology and a clinically adopted standard is primarily institutional rather than technical. Without a host institution, a sustained reviewer workforce, and regulatory recognition, the resource risks remaining an academic artifact rather than a clinical tool.
Keywords RadLex; radiology ontology; Ukrainian language; medical terminology; structured reporting; clinical natural language processing; low-resource localization; eHealth Ukraine; health informatics; European Health Data Space
References
1. Chepelev LL, Kwan D, Kahn CE, Filice RW, Wang KC. Ontologies in the new computational age of radiology: RadLex for semantics and interoperability in imaging workflows. RadioGraphics. 2023;43(3):e220098. https://doi.org/10.1148/rg.220098
2. Langlotz CP. RadLex: a new method for indexing online educational materials. RadioGraphics. 2006;26(6):1595–1597. https://doi.org/10.1148/rg.266065168
3. Wang KC. Standard lexicons, coding systems and ontologies for interoperability and semantic computation in imaging. J Digit Imaging. 2018;31(3):353–360. https://doi.org/10.1007/s10278-018-0069-8
4. Kuznietsova I, Kuznietsov D, et al. Insight into the digital health system of Ukraine (eHealth): trends, definitions, standards, and legislative revisions. Int J Telerehabil. 2023;15(2):e6549. https://doi.org/10.5195/ijt.2023.6549
5. Plis K, Kronfeld M, Banerjee I. Large-scale integration of DICOM metadata into HL7-FHIR for medical research. J Digit Imaging. 2025;38:30. https://doi.org/10.1007/s10278-025-01388-2
6. Filice RW, Kahn CE. Biomedical ontologies to guide AI development in radiology. J Digit Imaging. 2021;34(6):1331–1341. https://doi.org/10.1007/s10278-021-00527-1
7. Shevchenko Scientific Society, Medical Sciences. Sketches of Ukrainian medical terminology: origins, peculiarities and methods of application. Proceedings of the Shevchenko Scientific Society, Medical Sciences; 2020.
8. Cherednichenko O, Kanishcheva O. Collection and processing of a medical corpus in Ukrainian (UKRMED). In: CEUR Workshop Proceedings, Vol. 2604; 2020.
9. Cabinet of Ministers of Ukraine. Ukraine integrates into the European medical statistics system: MOH and OECD to exchange data for general medical reporting from 2026. Press release; 2024. https://www.kmu.gov.ua/
10. European Society of Radiology (ESR). ESR paper on structured reporting in radiology — update 2023. Insights Imaging. 2023;14:199. https://doi.org/10.1186/s13244-023-01560-0
11. Pinto dos Santos D, Hempel JM, Kloeckner R, et al. Structured reporting in radiology: a systematic review to explore its potential. Eur Radiol. 2022;32(4):2837–2854. https://doi.org/10.1007/s00330-021-08327-5
12. Marwede D, Daumke P, Marko K, et al. RadLex — deutsche Version eines Radiologie-Lexikons zur Indexierung von Bild- und Befundinformationen. Fortschr Röntgenstr (RöFo). 2009;181(1):38–44. https://doi.org/10.1055/s-2008-1027895
13. Pinto dos Santos D, et al. Structured reporting in radiology: German and European radiology societies’ point of view. Insights Imaging. 2021;12:117. https://doi.org/10.1186/s13244-021-01063-0
14. Pavlenko O. The future of Ukrainian healthcare: the digital opportunity. J Glob Health. 2025;15:03039. https://doi.org/10.7189/jogh.15.03039
15. Solventum/3M. Innovative approaches to ICD-11 implementation: a look at what’s working around the world. Inside Angle Blog; 2023. https://www.solventum.com/
16. World Health Organization. ICD-11 implementation. Geneva: WHO; 2024. https://www.who.int/standards/classifications/frequently-asked-questions/icd-11-implementation
17. Radiological Society of North America. RadLex Radiology Lexicon. https://www.rsna.org/practice-tools/data-tools-and-standards/radlex-radiology-lexicon
18. Bressem KK, Adams LC, Gaudin RA, et al. Highly accurate classification of chest radiographic reports using deep learning NLP. Bioinformatics. 2020;36(21):5107–5115. https://doi.org/10.1093/bioinformatics/btaa668
19. Marquis M, Bossenko I, Ross P. RadLex and SNOMED CT integration: a pilot study for standardising radiology classification. Insights Imaging. 2025;16:62. https://doi.org/10.1186/s13244-025-01935-5
20. Neri E, Coppola F, Larici AR, et al. Structured reporting of chest CT in COVID-19 pneumonia: a consensus proposal. Insights Imaging. 2020;11(1):92. https://doi.org/10.1186/s13244-020-00901-7
21. Casey A, Davidson E, Poon M, et al. A systematic review of natural language processing applied to radiology reports. BMC Med Inform Decis Mak. 2021;21:179. https://doi.org/10.1186/s12911-021-01533-7
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Oleksandr Berezovskyi
This work is licensed under a Creative Commons Attribution 4.0 International License.
This article is published open access under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence.
You are free to: share and adapt the work for any purpose, even commercially.
Conditions: provide appropriate credit to the authors and the Georgian Medical Journal (GMJ), link to the licence, and indicate if changes were made. Do not apply legal terms or technological measures that legally restrict others from doing anything the licence permits.
Licence: creativecommons.org/licenses/by/4.0 · Legal code: creativecommons.org/licenses/by/4.0/legalcode
Third-party material. Images or other content credited to a third party are not covered by CC BY 4.0; permission must be obtained from the rights holder for reuse beyond statutory exceptions.
Authors' rights. Authors retain copyright. First publication rights are granted to GMJ.
Data and code. Where provided, datasets or code may carry their own licences; please follow the licence stated in the article or repository record.
Listen on Spotify
Apple Podcasts
YouTube
Amazon Music
Castbox
Pocket Casts
LinkedIn
Facebook
X
Instagram
Substack
Bluesky
ResearchGate
