Typical values of z-resolution for different Digital Breast Tomosynthesis systems evaluated in a multicenter study.

PURPOSE: The aim of the present study, conducted by a working group of the Italian Association of Medical Physics (AIFM), was to define typical z-resolution values for different digital breast tomosynthesis (DBT) models to be used as a reference for quality control (QC). Currently, there are no typical values published in internationally agreed QC protocols. METHODS: To characterize the z-resolution of the DBT models, the full width at half maximum (FWHM) of the artifact spread function (ASF), a technical parameter that quantifies the signal intensity of a detail along reconstructed planes, was analyzed. Five different commercial phantoms, CIRS Model 011, CIRS Model 015, Modular DBT phantom, Pixmam 3-D, and Tomophan, were evaluated on reconstructed DBT images and 82 DBT systems (6 vendors, 9 models) in use at 39 centers in Italy were involved. RESULTS: The ASF was found to be dependent on the detail size, the DBT angular acquisition range, the reconstruction algorithm and applied image processing. In particular, a progressively greater signal spread was observed as the detail size increased and the acquisition angle decreased. However, a clear correlation between signal spread and angular range width was not observed due to the different signal reconstruction and image processing strategies implemented in the algorithms developed by the vendors studied. CONCLUSIONS: The analysis led to the identification of typical z-resolution values for different DBT model-phantom configurations that could be used as a reference during a QC program.

Grayscale calibration and quality assurance of diagnostic monitors in a PACS system.

PURPOSE: The aim of this study was to calibrate monitors used in soft-copy review of diagnostic images in a pictures archiving and communication system (PACS) and to assess critical quality assurance (QA) parameters through appropriate checks. MATERIALS AND METHODS: Barco [cathode ray tube (CRT) and liquid crystal display (LCD)] and EIZO (LCD) monitors were evaluated. Calibration and QA controls were carried out during acceptance tests on the systems and every 6 months according to the Task Group 18 (TG18) report by the American Association of Physicists in Medicine (AAPM). The parameters in question include: maximum luminance, contrast ratio, luminance response, spatial resolution and angular response. A subjective evaluation of image quality was also conducted by a number of radiologists. RESULTS: Barco medical monitors' results were well within tolerances, with significant parameters persisting over time. EIZO nonmedical monitors showed rapid performance deterioration below the minimum requirements. Calibration had to be repeated only in a few cases. Radiologists' evaluations showed that monitor quality is equal to or even better than that of conventional films. CONCLUSIONS: Medical monitors turned out to be fully adequate to the task. Periodic QA tests are, however, absolutely necessary to ensure constant quality levels.

[The evaluation of the physical characteristics of a volumetric computer tomograph]. / Valutazione delle caratteristiche fisiche di un tomografo computerizzato volumetrico.

Spiral or volumetric computed tomography (CT) is a new scanning technique which allows the scanning of body regions with a continuously rotating system based on the slip ring technology; the patient is also moved continuously, synchronously with data acquisition. The physical characteristics of spiral CT image acquisition were compared with those of conventional CT images. The modulation transfer function (MTF) has the same values for medium-resolution filters, but lower values for spiral CT for high-resolution and frequency-enhancement filters. The slice sensitivity profile (SSP) describes the longitudinal image resolution for multiplanar reconstructions and was measured in terms of FWHM of the SSP curve. We obtained, for 10-mm slice thickness, a FWHM = 10.4 mm (conventional CT), versus 10.7 mm (Spiral CT), while, for 5-mm slice thickness, the corresponding values were 5.2 mm (conventional CT) and 5.5 mm (spiral CT). Noise was evaluated simply by measuring the standard deviation of the CT numbers, in a region of interest, of a uniform image and with the power spectrum or Wiener spectrum of the same image. To assess overall image quality and yield, the noise equivalent quanta (NEQ) value was also calculated. The values were a little lower for the spiral technique, particularly with high-resolution and enhancement or convolution filters. Dosimetric evaluation of the computed tomography dose index (CTDI) and of the multiple scan average dose (MSAD) was done using an acquisition protocol for average lung dose, in an anthropomorphic phantom and with TL dosimeters. The MSAD was 6.17 +/- 0.20 cGy for conventional CT and 5.98 +/- 0.23 cGy for Spiral CT, while lung dose was 3.25 +/- 0.12 cGy and 3.01 +/- 0.16 cGy, respectively.

[The spatial resolution of the digital storage phosphor system. The monitor and film compared]. / La risoluzione spaziale del sistema digitale con fosfori a memoria. Monitor e pellicola a confronto.

The physical characteristics of radiographic images, namely spatial resolution and contrast, have obvious effects upon diagnostic image usefulness. We investigated the spatial resolution of both radiographs and magnified digital obtained with a storage phosphor system, in comparison with a film-screen combination. This study was carried out on the conventional radiographs of a phantom grid 0.5 mm thick, with resolution ranging from 0.5 to 10 lp/mm. Each examination was compared at naked eye and with the electronic evaluation of a region of interest on both standard and magnified views or by digitization with a charge coupled detector (CCD) television camera followed by the computing of the modulation transfer function curve. Our results demonstrate a higher spatial resolution of direct magnification, on both digital and film-screen pictures (over 5 lp/mm). On the contrary, the electronic magnification on the monitor yields the same spatial resolution as non-magnified digital images (up to 4.3 lp/mm). By selecting appropriate regions of interest, we could demonstrate the compression of the non-magnified images on the monitor. The modulation transfer curves show that direct magnification yields higher spatial resolution than electronic magnification and non-magnified views. Viewing electronically magnified images on the monitor yields the same resolution as contact radiographs: the monitor offers the advantage of an easier study of the regions of interest.

[Dosimetric evaluations in dental radiology: comparison of the digital system and the conventional system]. / Valutazioni dosimetriche nella radiologia odontoiatrica: sistema digitale con fosfori fotostimolabili e tradizionale a confronto.

The dosimetric values measured on digital dental examinations (panoramic radiography and cephalography) were compared with those obtained with screen-film combinations. The X-ray dose was measured at the critical organs (lens, thyroid, cervical cord) on an anthropomorphic phantom, using thermoluminescence dosimeters and an ionization chamber: the radiographic unit was set at the same exposure values used for standard human studies in adults and children. The mean thyroid dose for screen-film panoramic radiographs was 0.037 mGy in the child and 0.053 in the adult; the lens and the cord doses were 0.009-0.012 mGy and 0.096-0.135 mGy, respectively. Digital panoramic radiography gave the thyroid a dose of 0.013 mGy in the child and 0.016 mGy in the adult: such a marked dose reduction was observed for the lens and the cervical cord too (0.0044-0.0052 mGy and 0.035-0.042 mGy, respectively). With conventional film cephalography the thyroid was given a mean dose of 0.042-0.072 mGy according to age and to patient's size; the lens received higher doses (0.420-0.720 mGy). These values were markedly reduced with the digital technique; 0.032-0.034 mGy to the thyroid gland and 0.320-0.340 mGy to the lens; spinal cord doses became negligible with digital exposures (0.009-0.010 mGy). The overall analysis of dosimetric values demonstrated a mean dose reduction averaging 40-60% with respect to standard film exposure. This difference is more apparent in the adult, since the linear dose-response pattern of the digital system allows good images to be acquired in "thicker" people without markedly increasing exposure values.

[Radiographic enlargement in the study of hand diseases. Digital versus conventional techniques]. / L'ingrandimento radiografico nello studio delle malattie della mano. Tecniche digitale e tradizionale a confronto.

The electronic magnification of digital images was compared with direct digital magnification and with plain radiograph magnification. A whole-body computed radiographic system with photostimulable phosphor plates contained inside standard X-ray cassettes was used. The small bones of the hands and wrists of 18 patients with traumatic (10 cases) or degenerative (8 cases) bone conditions mainly due to chronic renal failure were studied. Each patient was examined with all three techniques: the images were retrospectively reviewed by four observers relative to resolution, contrast, visibility of the lesions and diagnostic value of each method. The statistical analysis of our results demonstrated better yield of direct digital magnification than of plain radiograph (p = 0.00043) and of electronic (p = 7.5 10(-13) magnification. This finding was mainly due to density and contrast optimization of digital images, in spite of their low spatial resolution. This feature allows good simultaneous representation of structures with different radiographic densities, as it happens in the hand and wrist. Electronic magnification yielded less brilliant results, even if compared with plain radiographic studies (p = 0.0032). However, this limitation was compensated for by the fact that this technique is easier and quicker to perform and that it does not require any additional X-ray exposure.

[The detectability of osteoarticular lesions of the extremities on the television monitor]. / Riconoscibilità delle lesioni osteoarticolari delle estremità al monitor televisivo.

The diagnostic accuracy of the light-box reading of conventional (film screen) radiographs of the extremities was compared with that of the same set of images displayed on a 1 k x 1 k interactive monitor after laser digitization. 389 alterations (23 nondisplaced fractures, 129 soft-tissue calcifications and 237 articular bone erosions), identified by two experienced radiologists on 66 conventional radiographs, were the reference standard. ROC statistical analysis was performed on 1,556 observations expressed by four readers. The overall diagnostic performance of the two display modalities were substantially equivalent: no statistically significant differences resulted on the whole, but two individual readers performed better with conventional images. No overall nor individual statistically significant difference was reobserved for the subset of articular erosions either. Light-box reading of conventional radiographs allowed a higher number of calcifications in the soft-tissues and of proximal (carpal) abnormalities to be detected. Although our results indicate the overall high fidelity of monitor-displayed laser-digitized images, major improvements in the performance of digital diagnostic workstations are still required before adopting monitors for routine radiologic activity.

[Detection of small bone lesions with digital radiography using storage phosphors]. / Riconoscibilità delle piccole lesioni scheletriche con radiografia digitale mediante fosfori a memoria.

To date, the skeletal imaging capabilities of digital radiography with storage phosphors have been poorly investigated, and the diagnostic accuracy of this technique has not been thoroughly assessed. To evaluate the performance of storage phosphor digital radiography we compared 66 conventional and 66 digital radiographs of small abnormalities of the extremities (fractures, erosions, calcifications). Conventional images were obtained with a low-speed screen-film system while digital ones were acquired with high-resolution (5 lp/mm max) phosphors and laser-printed on a 8" x 10" film. Two experienced radiologists defined the gold standard (389 abnormalities) and four radiologists scored the findings (1,556 observations) on a five-point discrete scale. ROC analysis indicated film and storage radiography to be equally effective in the overall detection of abnormalities. No difference was found in the individual performances of the four readers, in the site subclasses (wrist, hand), and in the specific detection of fractures and erosions. Digital radiography proved to be superior to conventional radiography in the detection of calcifications in all sites and particularly in the wrist (p less than 0.05). Storage phosphor radiography may replace conventional radiography of the extremities without causing any significant decrease in diagnostic accuracy.