RESUMO
PURPOSE: This investigation aims to study the optimization in digital mammography, considering a diverse percentage of breast glandularity using the contrast-detail metric. METHODS: The Figure of Merit (FOM), defined as the ratio of the square of the Inverted Image Quality Figure (IQFINV) by the Mean Glandular Dose (MGD), was used. A Monte Carlo simulation study was carried out to calculate the Normalized Glandular Dose (DgN). A contrast detail analysis employing the test object Contrast-Detail Mammography Phantom (CDMAM, type 3.4) was performed in the Hologic digital mammography system-model Selenia located in the Research Center in Radiation Sciences and Technologies (CPqCTR) facilities (Brazil). It employed the CIRS phantom with 20 %, 30 %, 50 % of glandularity, and 6.0 cm in thickness. RESULTS: It was obtained new acquisition parameters for all glandularities that achieved a decrease in the MGD up to â¼ 50 %, maintaining the same image quality. The study was validated using the CIRS, TORMAM, and ACR phantoms through the contrast-to-noise ratio (CNR), the signal-to-noise ratio (SNR), and the MGD values obtained with the optimized parameters and the four AEC modes, which are the optimization proposed by the manufacturer. CONCLUSIONS: In this work, a new procedure was proposed that estimated the IQFINV value using the equivalence criterion between the CIRS phantom and the CDMAM test object with their respective PMMA plates. Based on the optimization carried out in this investigation, the AEC parameters, considering diverse glandularities, could be improved. This achievement permits the implementation of new protocols that optimize the ratio between the image's quality and the breast dose with 6.0 cm in thickness and 20 %, 30 %, and 50 % glandularity using contrast-detail metric.
