[Phantom study for the detection of simulated lesions in five different digital and one conventional mammography system]

Rofo. 2004 Aug;176(8):1127-32. doi: 10.1055/s-2004-813261.
[Article in German]

Abstract

Purpose: Experimental phantom study for the detection of simulated lesions with five different digital and one conventional screen-film mammography system.

Materials and methods: Three radiographs were obtained at various configurations of the phantom with one conventional screen-film system (Mammomat 3000 N) (Siemens), five digital systems (high resolution computed radiography system [Fuji/Siemens], one a-Si detector [GE Medical Systems], two a-Se detectors [Siemens; Hologic / Lorad] and one CCD detector [Fischer Imaging]), applying the same exposure parameters. The Wisconsin Mammographic Random Phantom, Model 152 A, was used. Five investigators with different experience in mammography (3 months to more than 4 years) evaluated the 18 randomly selected radiographs.

Results: No significant differences were found in the detection rate of simulated breast lesions for conventional screen-film mammography (84.9 %), high resolution computed radiography (86.7 %) and digital mammography with an a-Si detector (89.8 %). Highly significantly better results (p < 0.001) were found with the two a-Se systems (97.3 %) and the CCD system (100 %).

Conclusion: Conventional screen-film mammography can be replaced by high resolution computed radiography and digital mammography with a-Si-, a-Se- and CCD-detectors. This has to be confirmed in further clinical studies.

Publication types

  • English Abstract

MeSH terms

  • Computer Simulation
  • Female
  • Humans
  • Mammography*
  • Observer Variation
  • Phantoms, Imaging*
  • Sensitivity and Specificity
  • Xeromammography*