Anonymization of DICOM electronic medical records for radiation therapy.

Electronic medical records (EMR) and treatment plans are used in research on patient outcomes and radiation effects. In many situations researchers must remove protected health information (PHI) from EMRs. The literature contains several studies describing the anonymization of generic Digital Imagin...

Descripción completa

Detalles Bibliográficos
Publicado en:Computers in Biology & Medicine Vol. 53; pp. 134 - 141
Autores principales: Newhauser, Wayne, Jones, Timothy, Swerdloff, Stuart, Newhauser, Warren, Cilia, Mark, Carver, Robert, Halloran, Andy, Zhang, Rui
Formato: Journal Article
Publicado: Elsevier B.V. Oct2014
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=110182569&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 110182569
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        00104825
        JC2
      jtl: Computers in Biology & Medicine
      issn: 00104825
      maglogo: N
    pubinfo:
      dt: Oct2014
      vid: 53
      pid: 82545
      pub: Elsevier B.V.
      place: Philadelphia, Pennsylvania
    artinfo:
      ui:
        110182569
        NLM25147130
        10.1016/j.compbiomed.2014.07.010
        NLM25147130
        PMC4170049
        110182569
      ppf: 134
      ppct: 7
      formats:
      tig:
        atl: Anonymization of DICOM electronic medical records for radiation therapy.
      aug:
        au:
          Newhauser, Wayne
          Jones, Timothy
          Swerdloff, Stuart
          Newhauser, Warren
          Cilia, Mark
          Carver, Robert
          Halloran, Andy
          Zhang, Rui
        affil: Department of Physics and Astronomy, Medical Physics Program, Louisiana State University, 202 Nicholson Hall, Baton Rouge, LA 70803, USA; Department of Medical Physics, Mary Bird Perkins Cancer Center, 4950 Essen Lane, Baton Rouge, LA 70809, USA
      sug:
        subj:
          Electronic Health Records
          Data Security
          Radiographic Image Enhancement
          Image Processing, Computer Assisted Methods
          Privacy and Confidentiality
          Reproducibility of Results
      ab: Electronic medical records (EMR) and treatment plans are used in research on patient outcomes and radiation effects. In many situations researchers must remove protected health information (PHI) from EMRs. The literature contains several studies describing the anonymization of generic Digital Imaging and Communication in Medicine (DICOM) files and DICOM image sets but no publications were found that discuss the anonymization of DICOM radiation therapy plans, a key component of an EMR in a cancer clinic. In addition to this we were unable to find a commercial software tool that met the minimum requirements for anonymization and preservation of data integrity for radiation therapy research. The purpose of this study was to develop a prototype software code to meet the requirements for the anonymization of radiation therapy treatment plans and to develop a way to validate that code and demonstrate that it properly anonymized treatment plans and preserved data integrity. We extended an open-source code to process all relevant PHI and to allow for the automatic anonymization of multiple EMRs. The prototype code successfully anonymized multiple treatment plans in less than 1min/patient. We also tested commercial optical character recognition (OCR) algorithms for the detection of burned-in text on the images, but they were unable to reliably recognize text. In addition, we developed and tested an image filtering algorithm that allowed us to isolate and redact alpha-numeric text from a test radiograph. Validation tests verified that PHI was anonymized and data integrity, such as the relationship between DICOM unique identifiers (UID) was preserved.
      pubtype: Academic Journal
      doctype: Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N