Stefano Rivetti, MSc, Azienda USL di Modena; Nico Lanconelli, PhD; Marco Bertolini, MSc; Reggio Emilia; Giovanni Borasi, MSc; Andrea Nitrosi, PhD; Domenico Acchiappati, PhD

Digital radiography systems are replacing films over a broad range of examinations. Given the severe demands of mammography for an imaging system, such as high spatial and contrast resolution, digital mammography commercial units appeared on the market only few years ago. Two main approaches have been developed in digital mammography: indirect conversion systems based on scintillator screens, and direct conversion systems based on a direct conversion of X-rays into electrical charges. Full Field Digital Mammography (FFDM) systems based on indirect-conversion detector generally have a poorer response, in terms of spatial resolution, with respect to detectors based on a direct conversion process. This is mainly caused by the blurring introduced by the phosphor, where X-rays are converted into light. This blurring could cause some problems in mammography, since high spatial-resolution properties are required here. The most common material used for direct conversion detectors for mammography is amorphous Selenium. Recently, various mammography systems based on amorphous Selenium have been developed and are currently available on the market.

Also recently, a novel approach for direct conversion detectors has been developed. Consequently, a new system for digital mammography following this approach will soon appear on the market. This system is manufactured by FujiFilm Medical Systems, and exploits a double layer of amorphous selenium, used as an X-ray sensor. The X-ray is converted into electrical signals in the first layer. Image electrical signals are efficiently read-out in the second layer, using an optical switch rather than an electrical switch. Through this, both miniaturization of pixel pitch and high S/N ratio are achieved. First, this should allow images with high spatial resolution to be received, as demanded by mammography. Second, by creating a highly pure accumulation layer of amorphous selenium, a high X-ray conversion rate is achieved.

In order to assess the performance of the new Fuji unit, we performed a complete characterization by measuring physical properties, such as Modulation Transfer Function (MTF), Noise Power Spectra (NPS), and Detection Quantum Efficiency (DQE). All measurements were obtained on a clinical system with no compressor and no anti-scatter grid. A low scatter condition is simulated with a 2 mm Al filter attached to the generator head, using a 28 kVp Mo-Mo beam (as recommended by IEC standard 61267 – radiation condition RQA-M2.

The response curve is determined by exposing the detector to a wide range of uniform X-ray exposures. Pre-sampling MTF is measured by using both the edge and the slit technique: an oversampled LSF is obtained in both cases and used for MTF calculation. The edge and the slit are oriented at a small angle (smaller than 5°) with respect to the pixel array. MTF is first calculated for horizontal and vertical axes, and then averaged for getting the final estimate. NPS is calculated from flat field images at different exposure levels. Four images are obtained for each exposure level; and for each image a fixed-size ROI is extracted, Fourier transformed and averaged, for getting the 2D NPS. The 1D NPS is then normalized for the squared mean signal value of the ROI, thus getting the Normalized NPS (NNPS). The product of NNPS and exposure (or air kerma) is independent of exposure for a linear, quantum-limited detector. Thus, we calculated this product for assessing the quantum-limited condition of the system. The same ROIs used for NNPS calculation were also used to perform the Relative Standard Deviation (RSD) analysis. RSD (i.e., standard deviation divided by average signal value) was calculated inside the ROI. RSD analysis allows a more complete investigation of the noise properties of the detector. In fact, the contributions of the quantum-statistical (Poisson) noise source, of a dose related (multiplicative) noise source, and of a dose independent (additive) noise source can be evaluated. For the determination of the X-ray photon fluence, HVL measurements are made with the 2 mm aluminum foil filtered beam. For all image acquisitions, the exposure to the detector was measured using a calibrated mammographic ionization chamber. The source-to-image distance is nearly 65 cm. The DQE is finally calculated from the estimated MTF and NNPS. Recently, our group developed a set of methods and informatics procedures for facilitating the image quality assessment of systems based on digital detectors. In this work, we applied and used these procedures for characterizing the novel system presented by Fuji.

This work has allowed us to assess correct implementation of the procedures developed by our group with a novel unit for digital mammography. The system analyzed is able to provide excellent properties, in terms of spatial resolution, thanks to the direct conversion of the X-rays. The final version of the paper will contain a more thorough investigation of the noise properties of the Fuji system and a complete physical characterization.

We are presenting the physical characterization of a novel system for digital mammography based on a amorphous Selenium detector. Preliminary results seem to demonstrate that this new apparatus is able to provide an image quality comparable to other systems available on the market.

International Electro Technical Commission, Medical diagnostic X-ray equipment. Radiation conditions for use in the determination of characteristics (IEC-61267, Geneva, Switzerland, 2003).

Rivetti S, Lanconelli N, Campanini R, et al. “Comparison of different commercial FFDM units by means of physical characterization and contrast-detail analysis.” Med. Phys. 33(11). 2006;4198-4209.

Nitrosi A, Bertolini B, Borasi B, et al. “Application of QC_DR Software for Acceptance Testing and Routine Quality Control of Direct Digital Radiography Systems: Initial Experiences using the Italian Association of Physicist in Medicine Quality Control Protocol,” to appear in Journal of Digital Imaging.