1 / 67

Transforming Clinical and Clinical Research Data to Ontology-Driven Linked Data

Transforming Clinical and Clinical Research Data to Ontology-Driven Linked Data. MATHIAS BROCHHAUSEN, SEPTEMBER 02, 2019 ONTOBRAS 2019 Porto Alegre, Brazil. Contents. Managing Data Using Ontologies Mapping Ontologies to Other Data Representations Collecting Data Immediately in RDF

tam
Download Presentation

Transforming Clinical and Clinical Research Data to Ontology-Driven Linked Data

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Transforming Clinical and Clinical Research Data to Ontology-Driven Linked Data MATHIAS BROCHHAUSEN, SEPTEMBER 02, 2019 ONTOBRAS 2019 Porto Alegre, Brazil

  2. Contents • Managing Data Using Ontologies • Mapping Ontologies to Other Data Representations • Collecting Data Immediately in RDF • Representational Accuracy for Informed Consent Data

  3. Managing Data Using Ontologies

  4. Using Inference to Fill Data Gaps All children living in smoking households x

  5. Using Ontologies to Manage Data Smoking household  household & has part SOME (bearer of smoker role) OWL FILE Data repository

  6. Positive Effects for Cohort Identification Smoking household  household & has part SOME (bearer of smoker role) Data repository All children living in smoking households?

  7. Data quality dimension from: DAMA UK Working Group: The six primary dimensions of data quality, 2013.

  8. Using inference to assist completeness Smoking household  household & has part SOME (bearer of smoker role) Data repository All children living in smoking households?

  9. Platform for Imaging in Precision Medicine • Enhancements to address challenges • Efficiency and sustainability • New tools for analysis • Improving ability to manage and analyze integrated datasets • Making clinical and non-image data more accessible and usable • Integrating image and non-image data • Supporting new data types Slide by Jonathan Bona. Used with permission.

  10. Slide by Jonathan Bona. Used with permission.

  11. Excerpt from Head and Neck Squamous Cell Carcinoma collection Grossberg  A, Mohamed A, Elhalawani H, Bennett W, Smith K, Nolan T, Chamchod S, Kantor M, Browne T, Hutcheson K, Gunn G, Garden A, Frank S, Rosenthal D, Freymann J, Fuller C.(2017).  Data from Head and Neck Cancer CT Atlas. The Cancer Imaging Archive.  DOI: 10.7937/K9/TCIA.2017.umz8dv6s Slide by Jonathan Bona. Used with permission.

  12. Excerpt from Head-Neck-PET_CT collection Martin Vallières, Emily Kay-Rivest, Léo Jean Perrin, Xavier Liem, Christophe Furstoss, Nader Khaouam, Phuc Félix Nguyen-Tan, Chang-Shu Wang, Khalil Sultanem. (2017). Data from Head-Neck-PET-CT. The Cancer Imaging Archive. doi: 10.7937/K9/TCIA.2017.8oje5q00 Slide by Jonathan Bona. Used with permission.

  13. Slide by Jonathan Bona. Used with permission.

  14. Approach • Transform these non-image data into shared, semantically-enhanced representations using Open Biomedical Ontologies Slide by Jonathan Bona. Used with permission.

  15. Disease and diagnosis for a lung cancer patient Slide by Jonathan Bona. Used with permission.

  16. Positive HPV status for a head and neck cancer patient Slide by Jonathan Bona. Used with permission.

  17. If our clinical data integration project interests you: Forthcoming: • Jonathan P. Bona, Tracy S. Nolan, and Mathias Brochhausen. "Ontology-Enhanced Representations of Non-image Data in The Cancer Imaging Archive". Proceedings of the International Conference on Biological Ontology 2018, Corvallis, OR, August 7 - 10, 2018. • https://www.researchgate.net/publication/326158241_Ontology-Enhanced_Representations_of_Non-image_Data_in_The_Cancer_Imaging_Archive

  18. Mapping Ontologies to Other Data Representations

  19. CDE and re-representation

  20. INGAGE

  21. Why not simple create an OWL class and map it to the CDE? • In order to provide computable semantics those OWL classes would need axiomatic definitions. • There would be a lot of fairly complex axioms if we add OWL classes for every primary and secondary use of the data. • We can using reasoning and create classes when we query the data instead.

  22. The proposal CDEs CDEs INGAGE INGAGE retrieves 1 to 1 mapping INGAGE Query Documented SPARQL queries INGAGE-OWL RDF data OWL classes

  23. MIABIS2.0 OBIB Mapping Table OBIB IRI RDF pattern MIABIS code OBIB label SPARQL query MIABIS label Only for 1-1 mappings For all other!

  24. Collecting Data Immediately in RDF

  25. Collecting RDF data • Can we build a tool that looks and behaves like a simple questionnaire tool (SurveyMonkey, etc.), but creates RDF data as participants fill in information? • Can we us the data participants provide for a real time preliminary comparison of their answers to others?

  26. The CAFÉ project https://cafe-trauma.com

  27. OOSTT • Ontology of Organizational Structures of Trauma centers and Trauma system • http://purl.obolibrary.org/obo/oostt.owl • funded by NIGMS (R01GM111324-01) • Creator: Mathias Brochhausen • Contributors: J. Ball, S.M. Bowman, W.R. Hogan, A. Hicks, J. Judkins, R.T. Maxson, R. Nabaweesi, J. Neil Otte, R. Pradhan, N.D. Sanddal, M.E. Tudoreanu, R.J. Winchell

  28. RDF-powered questionnaire • Utecht J, Judkins J, Colvin T Jr., Otte JN, Rogers N, Rose R, Alvi M, Hicks A, Ball J, Bowman SM, Maxson RT, Nabaweesi R, Pradhan R, Sanddal ND, Tudoreanu ME, Winchell R, Brochhausen M. OOSTT: a Resource for Analyzing the Organizational Structures of Trauma Centers and Trauma Systems. CEUR Workshop Proc. 2016 Aug;1747. http://ceur-ws.org/Vol-1747/IT504_ICBO2016.pdf

  29. RDF-powered questionnaire

  30. RDF data about trauma centers and trauma systems

  31. Compliance to ACS Criteria Heat maps of compliance for trauma program components to ACS standards for a Level 1 and a Level 2 trauma center.

  32. Building a Pattern Repository for Instantiation Patterns

  33. Representing a individual with HPV who has been diagnosed with HPV Slide by Jonathan Bona. Used with permission.

  34. Comparative Assessment for Environments of Trauma Care • CAFE aims to represent the organizational structures of trauma centers and trauma systems in RDF, its internal collection of 150 RDF instantiation patterns were built to model the organizations described by answers to survey questions. • CAFE’s many general instantiation patterns exist only in the project’s internal database with no easy way to reuse or share them. Slide by Jonathan Bona. Used with permission.

  35. What pattern should I use? • Even experienced ontology users will not necessarily decide on the exact same patterns to represent the same kind of instance data. • E.g. representing a patient and their prescription Slide by Jonathan Bona. Used with permission.

  36. A drug prescribing process that has a person bearing the patient role as its participant Slide by Jonathan Bona. Used with permission.

  37. A drug prescribing process that has a particular patient role as its participant, and the bearer of that role Slide by Jonathan Bona. Used with permission.

  38. A drug prescribing process that realizes a particular patient role, and the bearer of that role Slide by Jonathan Bona. Used with permission.

  39. A drug prescription that is the specified output of a drug prescribing process and is about a particular person … Slide by Jonathan Bona. Used with permission.

  40. Prescribing a drug to a patient • None of these patterns is glaringly wrong • Some might be preferred over others for certain uses • All other things being equal, I should prefer to use patterns that other people are already using / will use Slide by Jonathan Bona. Used with permission.

  41. Sharing ontology use patterns for instance data Slide by Jonathan Bona. Used with permission.

  42. Sharing ontology use patterns for instance data

  43. Towards better sharing • Build ontology use patterns repository • search, browse, view, download, ontology usage patterns. • Implementation • Python web application • Backed by triple store • Very small application ontology http://purl.org/ontology-use-patterns Slide by Jonathan Bona. Used with permission.

  44. Slide by Jonathan Bona. Used with permission.

  45. Slide by Jonathan Bona. Used with permission.

  46. Slide by Jonathan Bona. Used with permission.

  47. Slide by Jonathan Bona. Used with permission.

  48. Slide by Jonathan Bona. Used with permission.

More Related