Skip to main content

Table 11 Results of the “One Health” queries performed that intend to acquire validated knowledge about the diagnosis and management of well-known medical conditions – The SPARQL SELECT queries appear within the Additional file 4 and the description of the queries appear within the subsection “Extracting locality-based modules with SNOMED CT and enabling One Health queries” of the section Materials and methods

From: Exploring semantic deep learning for building reliable and reusable one health knowledge from PubMed systematic reviews and veterinary clinical notes

  1. Each query qiVU, with i = {1,2,3}, is the union of the results obtained for the query qiV over VetCN dataset and the query qiV over PMSB dataset (see Table 9 for details). The cells with grey background indicate that there are common UMLS Metathesaurus concept pairs in both VetCN and PMSB datasets, and therefore, the total number of results for the query qiVU is lower that the summation of the results obtained for the query qiV in each dataset (see the rows with grey background in Table 7 for details of the common UMLS Metathesaurus concept pairs retrieved). As more than one SNOMED CT concept can map one UMLS Metathesaurus concept, the number of results for the query qiVM is equal to or lower than the number of results for the query qiVS, with i = {1,2,3}. Each SPARQL SELECT query qiVR, with i = {1,2,3}, retrieves the asserted and inferred descendants (with FaCT++) of those SNOMED CT concepts mapped to candidate concepts of the SNOMED CT pairs retrieved from the SPARQL SELECT query qiVS