Image quality assessment using the CD-DISC phantom for vascular radiology and vascular surgery

Lara Struelens, Lionel Hambach, Nico Buls, Kristien Smans, Françoise Malchair, Marie-Thérèse Hoornaert, Filip Vanhavere, Hilde Bosmans, Emiliano D'Agostino

    Research outputpeer-review

    Abstract

    The purpose of the study was to evaluate image quality (IQ) associated with vascular radiology and vascular surgery procedures in Belgium and to determine reference values for future image quality assessment. IQ was evaluated with the CD-DISC contrast-detail phantom. Images of the phantom were acquired for both fluoroscopy and subtraction images on 21 systems. Three observers evaluated the images by determining the threshold contrast visible for every diameter. We observed a large difference in IQ between the centres. No straightforward correlation could be found with radiation dose or other exposure settings. A comparison was made with the image quality evaluation of the systems performed with the TOR[18FG] phantom for fluoroscopy. There is no clear correlation observed between the results of the CD-DISC phantom and the TOR phantom. However, systems with very poor or very good image quality could be detected by both phantoms. An important result is that a 75th percentile reference contrast-detail curve could be proposed to separate the best centres from these with poorer quality. We introduced also a 25th percentile. Centres with IQ above this value are recommended to lower the dose and work with acceptable rather than excellent image quality.
    Original languageEnglish
    Pages (from-to)348-356
    JournalEuropean Journal of Radiology
    Volume67
    Issue number2
    DOIs
    StatePublished - Jul 2008

    Cite this