A national approach for automated collection of standardized and population-based radiation therapy data in Sweden

Research output: Contribution to journalArticle

Abstract

Purpose To develop an infrastructure for structured and automated collection of interoperable radiation therapy (RT) data into a national clinical quality registry. Materials and methods The present study was initiated in 2012 with the participation of seven of the 15 hospital departments delivering RT in Sweden. A national RT nomenclature and a database for structured unified storage of RT data at each site (Medical Information Quality Archive, MIQA) have been developed. Aggregated data from the MIQA databases are sent to a national RT registry located on the same IT platform (INCA) as the national clinical cancer registries. Results The suggested naming convention has to date been integrated into the clinical workflow at 12 of 15 sites, and MIQA is installed at six of these. Involvement of the remaining 3/15 RT departments is ongoing, and they are expected to be part of the infrastructure by 2016. RT data collection from ARIA®, Mosaiq®, Eclipse™, and Oncentra® is supported. Manual curation of RT-structure information is needed for approximately 10% of target volumes, but rarely for normal tissue structures, demonstrating a good compliance to the RT nomenclature. Aggregated dose/volume descriptors are calculated based on the information in MIQA and sent to INCA using a dedicated service (MIQA2INCA). Correct linkage of data for each patient to the clinical cancer registries on the INCA platform is assured by the unique Swedish personal identity number. Conclusions An infrastructure for structured and automated prospective collection of syntactically interoperable RT data into a national clinical quality registry for RT data is under implementation. Future developments include adapting MIQA to other treatment modalities (e.g. proton therapy and brachytherapy) and finding strategies to harmonize structure delineations. How the RT registry should comply with domain-specific ontologies such as the Radiation Oncology Ontology (ROO) is under discussion.

Details

Authors
  • Tufve Nyholm
  • Caroline Olsson
  • Måns Agrup
  • Peter Björk
  • Thomas Björk-Eriksson
  • Giovanna Gagliardi
  • Hanne Grinaker
  • Adalsteinn Gunnlaugsson
  • Anders Gustafsson
  • Magnus Gustafsson
  • Bengt Johansson
  • Stefan Johnsson
  • Magnus Karlsson
  • Leif Nyström
  • Eva Onjukka
  • Johan Reizenstein
  • Johan Skönevik
  • Karin Söderström
  • Alexander Valdman
  • Björn Zackrisson
  • Anders Montelius
Organisations
External organisations
  • Uppsala University
  • Linköping University Hospital
  • Mälar Hospital
  • Sahlgrenska University Hospital
  • Karolinska University Hospital
  • Swedish Radiation Safety Authority
  • Cureos AB
  • Örebro University Hospital
  • Kalmar County Hospital
  • Umeå University
  • Sahlgrenska Academy
  • Regional Cancer Center Western Sweden
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Cancer and Oncology
  • Radiology, Nuclear Medicine and Medical Imaging

Keywords

  • Data integration, Data standardization, Medical informatics, Radiation Oncology informatics, Radiation therapy data
Original languageEnglish
Pages (from-to)344-350
Number of pages7
JournalRadiotherapy and Oncology
Volume119
Issue number2
Publication statusPublished - 2016 May 1
Publication categoryResearch
Peer-reviewedYes