Dosimetry and image quality in digital mammography

The mammography is the most effective diagnostic method for breast disease detection. Digital systems improve image quality, but not always follow optimization protocols for image acquisition and it may result in higher doses absorbed by the mammary glands. Therefore, it is important to conduct radiological protection studies in patients. Thus, this study aims to estimate the mean glandular dose (MGD) in patients who underwent mammograms in digital systems in Recife, Brazil, and to analyze the image quality. This study was conducted with: one Siemens mammography unit and a Carestream CR reader, Kodak DirectView; two Lorad Hologic Selenia digital mammography DR, and one Lorad M-IV analogic equipment with a Kodak film processor, X-Omat 2000. In order to evaluate the MGD, tests were conducted in patients with ages between 40 and 64 years old and breast thickness between 2 and 9 cm. Only the routine incidents were taken into account (RCC, LCC, RMLO, LMLO). The irradiation parameters of 5200 mammograms were analyzed among these four equipment. were recorded (kV, mAs, anode/filter), as well as the thickness of the compressed breast and the patient's age. The MGD was calculated using the methodology described by Dance et al. The image quality was assessed through the modulation transfer function (MTF), the signal to noise ratio (SNR) and contrast to noise ratio (CNR) for different thicknesses of phatom, according to Humam Health Series 17 document of the International Atomic Energy Agency. Furthermore, the quality of 50 mammograms were evaluated by radiologists. The results showed that the value of MGD was 2.70±0.50 mGy to screen-film equipment, 3.91±0.72 mGy for CR equipment, 2.86±0.53 mGy for the DR1 and 3.83±0.70 mGy for the DR2. The average thicknesses were, respectively: 4.82 cm; 6.73 cm; 6.02 cm; 6.11 cm. The compression force was smaller than indicated for the digital equipment and greater for screen-film. The MGD values obtained with the phantom were lower than those obtained for patients with breast equivalent thickness. The SNR was suitable for all equipment. In most thicker breasts, CNR was not satisfactory, but DGM was below the recommended range for this type of breast, allowing to conclude that an adjustment in automatic exposure control is needed to improve the CNR, even though MGD might increase. The MTF was satisfactory for both CR and DR1 equipment. The evaluation of clinical images showed failure in positioning which indicates lack of training of mammography technicians. In general this study expects to awake and educate professionals of this field about the importance of radiation protection concepts in digital mammograms. (author)