52 SFPM working group on DBT internal quality control. (December 2018)

• Record Type: Journal Article
• Authors: Voyeau-Gautier, M.; Villani, N.; Guillerminet, C.; Michel, C.; Sage, J.
• Is Part Of: Physica medica. Volume 56(2018)Supplement 1
• Page Start: 29
• Publication Date: 2018-12
• Subjects: 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.2018.09.065
• Languages: English
• ISSNs: 1120-1797

Abstract

Abstract : Introduction: The introduction of Digital Breast Tomosynthesis (DBT) within the French breast cancer screening program is forecast by the authorities in the near future. A regulatory internal and external quality control is an essential prerequisite to this introduction. The French Society for Medical Physics (SFPM) has appointed a working group in order to establish recommendations in the scope of DBT and synthetic_2D internal quality control. These recommendations need to stand for the specific French context, in France, mammography is essentially done in private centers that do not necessarily have competence in medical physics. Internal quality controls are therefore principally intended to be done by technicians under radiologist responsibility. Methods: Literature has been reviewed and a feasibility study has been done. Then several metrics have been tested on a weekly basis for several months: acquisition parameters consistency, image score, SDNR, SNR, homogeneity and volume viewer lag test. These tests are consistent with the French regulatory internal quality controls defined for full field digital mammography (2D) and complement DBT quality controls recommended by EUREF[1] . Four phantoms have been evaluated: MTM 100, TOMOMAM, ACR 086 and GAMMEX DBT Phantom. Tests have been done on 3 systems from different brands: GEHC, HOLOGIC and SIEMENS. Results: Each week, half an hour is needed for the tests. The results show that MTM 100, ACR 086 and GAMMEX DBT Phantoms are not discriminant. The visual method to evaluate image score shows an important variability of the results for 2D, synthetic_2D as well as DBT with coefficients of variation in the range of 0.1 to 0.4. Some of the metrics (SNR, SDNR, homogeneity) make it possible to detect system performance degradation. The volume viewer lag test has been abandoned due to technical constraints. Conclusions: The tests requiring a short time and enabling to detect performance degradation are suitable to routine implementation in medical imaging centers. The study should be extended to all brands of DBT systems in order to refine the scope and methodology of internal quality controls and to establish acceptability criteria. The assessment of the phantoms needs to be deepened in order to stabilize the interpretation grid for visual image score analysis.