Medical Knowledge Resources for Text-Mining of Health Records in Czech, Polish, and Slovak
| Authors | |
|---|---|
| Year of publication | 2022 |
| Type | Article in Proceedings |
| Conference | Proceedings of the Sixteenth Workshop on Recent Advances in Slavonic Natural Languages Processing, RASLAN 2022 |
| MU Faculty or unit | |
| Citation | |
| web | |
| Keywords | EHR; electronic health records; healthcare text; UMLS; ICD10; SNOMED CT; MedDRA; MeSH; NLP; natural language processing; Slavic languages; Polish; Czech; Slovak |
| Description | Knowledge extraction from medical text in small languages like Czech, Polish or Slovak is challenging due to the insufficiency of languagespecific medical resources (pretrained models, ontologies, dictionaries). This paper is a survey of noteworthy options for researchers targeting these languages, divided into two sections. First, since the UMLS Metathesaurus for English is by far the most extensive and detailed medical knowledge resource in Western medicine, appreciable results can be achieved by machine-translating the mined text to English – therefore, the relevant English components of UMLS are introduced. Second come the languagespecific resources for each language, detailing the publishing institutions, current website locations, contents, and file formats. The contribution of this paper is in collecting and pre-screening widely disparate sources needed for successful medical knowledge extraction in Central European Slavic languages. |
| Related projects: |