Experimental evaluation of seven quality control phantoms for digital breast tomosynthesis. (January 2019)

• Record Type: Journal Article
• Authors: Sage, Julie; Fezzani, Karen L.; Fitton, Isabelle; Hadid, Lama; Moussier, Aurélie; Pierrat, Noëlle; Martineau, Antoine; Dreuil, Serge; Heulers, Loïc; Etard, Cécile
• Is Part Of: Physica medica. Volume 57(2019)
• Page Start: 137
• Page End: 144
• Publication Date: 2019-01
• Subjects: ASF artefact spread function -- DBT digital breast tomosynthesis -- EUREF European Reference Organisation for Quality Assured Breast Screening and Diagnostic Services -- FWHM full width at half-maximum -- GEHC general electric health care -- SSP slice sensitivity profile; Breast -- Diagnostic imaging -- Digital breast tomosynthesis -- Quality control -- Screening; Medical physics -- Periodicals; Biophysics -- Periodicals; Biophysics -- Periodicals; Imagerie médicale -- Périodiques; Radiothérapie -- Périodiques; Rayons X -- Sécurité -- Mesures -- Périodiques; Physique -- Périodiques; Médecine -- Périodiques; 610.153
• DOI: 10.1016/j.ejmp.2018.12.031
• Languages: English
• ISSNs: 1120-1797

Abstract

Highlights: Visual image score analysis is recommended to check the whole DBT chain quality. ACR, TOMOMAM®, TOMOPHAN® appeared to be pertinent for evaluation of DBT image score. Phantom's manufacturers should collaborate with DBT system's manufacturers. Abstract: Purposes: The introduction of digital breast tomosynthesis (DBT) into the French breast cancer screening program is forecast by the authorities. The aim of the present study was to evaluate image quality phantoms to be used as internal quality controls. Methods: Seven breast phantoms dedicated to quality control in mammography were evaluated on reconstructed DBT images: ACR Model 015, BR3D, DBT QC model 021, Mam/Digi-EPQC, MTM100, TOMOMAM® and TOMOPHAN®. Two representative image parameters of DBT images were studied: image score and z-resolution, when inserts were included in the phantom, on five DBT systems of three different brands. Three observers were involved. Results: The MTM100, Mam/Digi-EPQC, BR3D, DBT QC model 021 phantoms' images presented artefacts affecting the image score. The ACR Model 015, TOMOMAM® and TOMOPHAN® phantoms appeared to be pertinent for DBT image score analysis. Due to saturation artefacts, Z-resolution results were not coherent with the theory for all phantoms except by using aluminium beads in the TOMOMAM® phantom. Conclusions: Phantom manufacturers should be encouraged to collaborate with DBT system manufacturers in order to design universal phantoms suitable for all systems for more complete quality control. From our study we can propose several specifications for an ideal and universal phantom designed for internal quality control in DBT. Phantoms should allow sensitive image score measurements. The background structure should be realistic to avoid artefacts. Phantoms should have a standard breast-like shape and size.