Evaluation of equipment performance, patient dose, imaging quality, and diagnostic coincidence in five Mexico City mammography services

Background. Regulations concerning the use of x-rays in medical diagnoses were published in Mexico in 1997. In this work, we evaluate technical aspects of mammography services in the Mexico City area and radiation dose and coincidence between the radiological interpretation by the institution radiologist and by a panel of experts. Methods. Following methodology proposed by the American College of Radiology and the European Community among others, we have evaluated the performance of six mammography systems in Mexico City public and private services. The studied services carry out approximately one half of the mammography studies in the capital's metropolitan area. Results. The systems comply with 53-82% of a total of 31 applied quality control tests and measurements, which include the mammography unit, x-ray generation, collimation, automatic exposure control, compression devices, grid and image receptor, film processing, darkroom, viewboxes, dose, film rejection, and image quality. The elements that most frequently fail are film processing, darkroom, and light boxes; average ACR phantom score is 11.2 (9.5, 12.0); mean average glandular dose measured with the phantom is 1.00 (0.71-1.15) mGy, and measured in patients is 1.75 (0.3, 4.9) mGy; coincidence between radiologic reports (BI-RADS) by the institution radiologist and a panel of experts is obtained in 35% of studied cases. Conclusions. Statistical analysis of results indicated that the level of equipment performance is correlated with image quality, image quality estimated by the panel of radiologists is correlated with phantom score, and coincidence in clinical mammography reports is not correlated with equipment performance and appears to depend on the radiologist's experience. (C) 2004 IMSS. Published by Elsevier Inc.