Proceedings of International Conference on Applied Innovation in IT
2020/03/10, Volume 8, Issue 1, pp.71-76
Concept Map for Clinical Recommendations Data and Knowledge Structuring
Giyzel Shakhmametova, Nafisa Yusupova, Rustem Zulkarneev, Yevgeniy Khudoba1
Abstract: The article deals with the problem of structuring medical texts of clinical recommendations, which are unstructured texts. A review of existing solutions in the field of analysis of unstructured texts of both non-specialized and medical nature was carried out, shortcomings of existing developments were identified, the need for a new software solution for structuring clinical recommendations was revealed, which, in turn, is demanded in clinical decision support systems. The method of structuring data and knowledge of clinical recommendations is described, as well as the general structure of the solution, as along with the process of forming a map of concepts, including graphematic, morphological, syntactic and semantic analysis of text. In conclusion, the results of implementation in the form of concept map fragments are presented, on the basis of which further product rules are formed, which are suitable for use in knowledge bases. The method is universal and can be applied to any clinical recommendations texts.
Keywords: Structuring Data and Knowledge, Unstructured Text, Clinical Recommendations, Concept Map, Production Rules
- R. Grishman, "Twenty-five years of informationextraction". Natural Language Engineering,vol. 25(6), 2019, pp. 677-692, doi: 10.1017/S1351324919000512.
- S. Grimes, "A Brief History of Text Analytics". EyeNetwork. Retrieved, June 2016.
- B. Presannan, N. Ramasubramanian andA.S. Vijayan, “Disease risk prediction from clinicaltexts”, 2020, doi:10.1007/978-981-32-9515-5_30.
- F. Dhombres, J. Charlet and Section Editors for theIMIA Yearbook Section on KnowledgeRepresentation and Management. Formal medicalknowledge representation supports deep learningalgorithms, bioinformatics pipelines, genomics dataanalysis, and big data processes. Yearbook ofMedical Informatics, vol. 28 (1), 2019, pp. 152-155, doi: 10.1055/s-0039-1677933.
- B. Séroussi, L.F. Soualmia and J.H. Holmes,Transforming data into knowledge: How to improvethe efficiency of clinical care? Yearbook of MedicalInformatics, vol. 26(1), 2017, pp. 4-6, doi:10.1055/s-0038-1637768.
- C. Combi and G. Pozzi. "Clinical informationsystems and artificial intelligence: Recent research
- trends". Yearbook of Medical Informatics, vol. 28(1), 2019, pp. 83-94, doi:10.1055/s-0039-16779156.
- Text Mining Software, SAS Text Miner | SAS.Renewal date: 18.03.2019. [Online]. Available:https://www.sas.com/en_us/ software/text-miner.html.
- General Architecture for Text Engineering. Renewaldate: 18.03.2019. [Online]. Available:https://gate.ac.uk.
- STATISTICA Text Miner. Renewal date:18.11.2019. [Online]. Available:http://statsoft.ru/products/STATISTICA_Data_Miner/STATISTICA_Text_Miner.
- Natural Language Toolkit NLTK 3.4 documentation.Renewal date: 18.03.2019. [Online]. Available:https://www.nltk.org.
- Unified Medical Language System (UMLS).Renewal date: 18.11.2019. [Online]. Available:https://www.nlm.nih.gov/ research/umls.
- MedLEE | MedLingMap. Renewal date: 18.11.2019.[Online]. Available: http://www.medlingmap.org/taxonomy/term/80.
- Apache cTAKES - clinical Text Analysis KnowledgeExtraction System. Renewal date: 18.11.2019.[Online]. Available: http://ctakes.apache.org.
- A. Maedche and S. Staab, "The TEXT-TO-ONTOOntology Learning Environment" (PDF).ResearchGate, July 2000.
- Dresden Ontology Generator for Directed AcyclicGraphs (DOG4DAG). Renewal date: 30.10.2019.[Online]. Available: http://www.biotec.tu-dresden.de/research/schroeder/dog4dag.