Skip to Main content Skip to Navigation
Master Thesis

Évaluation de la concordance dans les études d'imagerie diagnostique : une étude de la qualité des données publiées

Abstract : Background: The relevance of imaging procedures may notably vary according to clinical setting which calls for valid evaluation methods applicable to diverse clinical settings. The problem of reproducibility of the results concerns also the diagnostic imaging studies. Key issues center on substitution potential for novel radiopharmaceuticals, or easier to use medical devices vs. reference diagnostic tools. A clearly defined step wise methodology is required to plan the study and adequate statistical tests are required to assess the agreement between both methods. The objectives of this study is to assess the quality and the relevance of statistical analyses in reports of recently published diagnostic medical imaging trials and to compare the results obtained by a junior doctor and two senior doctors. Methods: We conducted a systematic review of literature in the MEDLINE database (via PubMed) to identify reports from 8 core journals (4 of nuclear medicine and 4 of radiology) with high impact factor, indexed over the past ten years. We have screened all potentially relevant articles with key words for an exhaustive search and assessed them to have final eligible articles which had been all full-text analyzed. A standardized data-extraction form was generated which comprised 29 items related to the study quality. We tried to identify all sources of methodological biases and missing data that could influence the interpretation of results. All selected articles were analyzed by 3 independent reviewers (1 junior and 2 seniors). Relevant items will be presented. Results: Eighty articles have been full-text analyzed. Twelve items are deserved being discussed. Particularly five of them because more than 90% of publications had these 5 items missing: to have “agreement” or “comparison” in their key words, to mention if the assessors belong to the same center or not, to give a justification of the sample size, to report if there were protocol deviations, and to discuss the clinical relevance of interpretation. There is a good agreement between junior and senior. Conclusions: Our findings show that several points could be improved in the reporting of diagnostic imaging trials to ensure the reliability of the conclusions.
Document type :
Master Thesis
Complete list of metadata

Cited literature [65 references]  Display  Hide  Download
Contributor : Université Paris Descartes - Scd <>
Submitted on : Wednesday, October 28, 2015 - 10:53:33 AM
Last modification on : Friday, October 23, 2020 - 4:42:47 PM
Long-term archiving on: : Friday, January 29, 2016 - 1:13:31 PM


  • HAL Id : dumas-01221504, version 1


Jules Tyaniu Zhang-Yin. Évaluation de la concordance dans les études d'imagerie diagnostique : une étude de la qualité des données publiées. Médecine humaine et pathologie. 2015. ⟨dumas-01221504⟩



Record views


Files downloads