Retrospective Intersectional Corpuslinguistic Analysis of Radiology Reports of Innsbruck Medical University (MedCorpInn)

Förderstelle: ÖAW Go!digital next Generation
Projektleitung:
  • Ass.Prof. Mag. Dr. Claudia Posch, Inst. f. Sprachen und Literaturen: Sprachwissenschaft, LFUI
  • Karoline Irschara, MA , Inst. f. Sprachen und Literaturen: Sprachwissenschaft, LFUI
  • Dr. med. Stephanie Mangesius, PhD
  • Dr. med. univ. Leonhard Gruber, PhD
  • Assoz.-Prof. Priv.-Doz. Dr.med. Bernhard Glodny, Facharzt für Radiologie, Universitätsklinik für Neuroradiologie
Mitarbeiterin:

Karoline Irschara, M.A., B.A., Inst. f. Sprachen und Literaturen: Sprachwissenschaft, LFUI

N.N.

N.N.

Partners and collaborators:

Medical University of Innsbruck:

  • Dr. med. univ. Tanja Djurdjevic, Fachärztin für Radiologie mit Schwerpunkt Neuroradiologie
  • Assoz.-Prof. Priv.-Doz. Dr. Astrid E. Grams, Fachärztin für Radiologie mit Schwerpunkt Neuroradiologie, Universitätsklinik für Neuroradiologie
  • Dr. med. univ. Leonhard Gruber, PhD, Facharzt für Radiologie, Universitätsklinik für Radiologie
  • Univ.-Prof.in Dr.in med. univ. Margarethe Hochleitner, Professorin für Gender Medizin, Direktorin des Frauengesundheitsbüros, Fachärztin für Innere Medizin
  • Drin. Anna Luger, Universitätsklinik für Radiologie

University of Cambridge:

Dr. Rafael Rehwald, Department of Radiology, University of Cambridge

University of Innsbruck

  • Mag. Dr. Heike Ortner, Institut für Germanistik
  • Mag. Dr. Gerhard Rampl, Sprachwissenschaft, Sprachen und Literaturen
  • Mag. Dr. Michael Tschuggnall, DBIS, Institut für Informatik
  • Mag. Dr. Eva Zangerle, DBIS, Institut für Informatik

No activity in healthcare is possible without language: All the settings in healthcare rely on different forms of communication. The project MedCorpInn is looking for patterns of language use connected to biases by investigating a large data set with methods of Corpus Linguistics of 100.000 pre-anonymized radiology reports from the Innsbruck University hospital. 
 
The project firstly will technically improve an existing corpus. Necessary steps are the enhancement of the extensive metadata (age, gender, type of insurance, mode of examination etc.) as well as measures for automated data processing and for further anonymization. Moreover, the existing part-of-speech annotation will be improved. The resulting corpus will, for example, be used to study if differences of language use (if any) are connected to social factors. Furthermore, pressing gender-medical questions can be examined within the corpus, e.g. if medical procedures are linked with certain social factors, or if there are gender specific differences regarding the precision of certain measurements (e.g. of lengths/diameters of organs, tumours or injuries).