MR-CT multi-atlas registration guided by fully automated brain structure segmentation with CNNs.

PURPOSE: Computed tomography (CT) is widely used to identify anomalies in brain tissues because their localization is important for diagnosis and therapy planning. Due to the insufficient soft tissue contrast of CT, the division of the brain into anatomical meaningful regions is challenging and is commonly done with magnetic resonance imaging (MRI). METHODS: We propose a multi-atlas registration approach to propagate anatomical information from a standard MRI brain atlas to CT scans. This translation will enable a detailed automated reporting of brain CT exams. We utilize masks of the lateral ventricles and the brain volume of CT images as adjuvant input to guide the registration process. Besides using manual annotations to test the registration in a first step, we then verify that convolutional neural networks (CNNs) are a reliable solution for automatically segmenting structures to enhance the registration process. RESULTS: The registration method obtains mean Dice values of 0.92 and 0.99 in brain ventricles and parenchyma on 22 healthy test cases when using manually segmented structures as guidance. When guiding with automatically segmented structures, the mean Dice values are 0.87 and 0.98, respectively. CONCLUSION: Our registration approach is a fully automated solution to register MRI atlas images to CT scans and thus obtain detailed anatomical information. The proposed CNN segmentation method can be used to obtain masks of ventricles and brain volume which guide the registration.

Correlation between enhancement characteristics of MR mammography and capillary density of breast lesions.

OBJECTIVE: To correlate capillary density of breast lesions using the markers D2-40, CD31, and CD34 with early and late enhancement of magnetic resonance mammography (MRM). MATERIALS AND METHODS: The local ethics committee approved this study, and informed consent was avail-able from all patients. The study included 64 women with 66 histologically proven breast lesions (41 malignant, 25 benign). MR-enhancement 1 min after contrast medium administration was determined in the tumor (It1/It0 ratio) and in comparison to the surrounding tissue (It1/It1-fat ratio). Capillary density was quantified based on immunohistological staining with D2-40, CD31, and CD34 in breast tumors and surrounding breast tissue. Mean capillary densities were correlated with contrast enhancement in the tumor and surrounding breast tissue. The Kruskal-Wallis test was used to test whether lesions with different MR enhancement patterns differed in terms of capillary density. RESULTS: For CD34, there was statistically significant correlation between capillary density and tumor enhancement (r = 0.329, p = 0.012), however not for the malignant or benign groups separately. Mean vessel number identified by staining with D2-40 and CD31 did not correlate significantly with tumor enhancement (D2-40: r = -0.188, p = 0.130; CD31: r = 0.095, p = 0.448). There were no statistically significant differences in capillary density between breast lesions with delayed enhancement or a plateau and lesions showing washout (Kruskal-Wallis test. D2-40: p = 0.173; CD31: p = 0.647; CD34: p = 0.515). CONCLUSION: Of the three markers tested, CD34 showed best correlation between early contrast enhancement on MRM and capillary density. Further studies are necessary to clearly demonstrate an association between capillary density and contrast enhancement in breast tumors and surrounding tissue.

Correlation of contrast agent kinetics between iodinated contrast-enhanced spectral tomosynthesis and gadolinium-enhanced MRI of breast lesions.

OBJECTIVES: Assessment of contrast agent kinetics in contrast-enhanced MRI (CE-MRI) with gadolinium-containing contrast agents offers the opportunity to predict breast lesion malignancy. The goal of our study was to determine if similar patterns exist for spectral contrast-enhanced digital breast tomosynthesis (CE-DBT) using an iodinated contrast agent. METHODS: The protocol of our prospective study was approved by the relevant institutional review board and the German Federal Office for Radiation Protection. All patients provided written informed consent. We included 21 women with a mean age of 62.4 years. All underwent ultrasound-guided biopsy of a suspect breast lesion, spectral CE-DBT and CE-MRI. For every breast lesion, contrast agent kinetics was assessed by signal intensity-time curves for spectral CE-DBT and CE-MRI. Statistical comparison used Cohen's kappa and Spearman's rho test. RESULTS: Spearman's rho of 0.49 showed significant (P = 0.036) correlation regarding the contrast agent kinetics in signal intensity-time curves for spectral CE-DBT and CE-MRI. Cohen's kappa indicated moderate agreement (kappa = 0.438). CONCLUSION: There is a statistically significant correlation between contrast agent kinetics in the signal intensity-time curves for spectral CE-DBT and CE-MRI. Observing intralesional contrast agent kinetics in spectral CE-DBT may aid evaluation of malignant breast lesions. KEY POINTS: • Contrast agent kinetics can be assessed using spectral digital breast tomosynthesis (DBT). • Contrast agent kinetics patterns in spectral DBT are similar to those in contrast-enhanced MRI. • Multiple contrast enhancement for spectral DBT gives additional diagnostic information.

Measurement of cerebral circulation times using dynamic whole-brain CT-angiography: feasibility and initial experience.

The objective of the study was to use 320-detector row 4D CT angiography (CTA) for measuring cerebral circulation times (CCT) and to assess early venous drainage (EVD) and shortening of CCT in arteriovenous malformations (AVM) and to compare with DSA. CCT of 12 physiological patients and five AVM patients were acquired using a 4D CTA protocol by recording cerebrovascular bolus passage time. In the AVM patients EVD time (EVDT) was measured. Identical measurements were performed on DSA for the AVM patients. It was found that the physiological CCTs were 5.8 ± 1.4 s (M ± SD). EVD was seen in all AVMs and resulted in a shortened CCT of 3.4 ± 1.1 s (p = 0.01). There was no significant difference for CCT and EVDT values derived from DSA and 4D CTA. Thus, the CCTs can be measured non-invasively using clinical 4D CTA. Early venous drainage with shortened CCTs was observed by 4D CTA in all five patients with AVMs.

Development of low-dose photon-counting contrast-enhanced tomosynthesis with spectral imaging.

PURPOSE: To demonstrate the feasibility of low-dose photon-counting tomosynthesis in combination with a contrast agent (contrast material-enhanced tomographic mammography) for the differentiation of breast cancer. MATERIALS AND METHODS: All studies were approved by the institutional review board, and all patients provided written informed consent. A phantom model with wells of iodinated contrast material (3 mg of iodine per milliliter) 1, 2, 5, 10, and 15 mm in diameter was assessed. Nine patients with malignant lesions and one with a high-risk lesion (atypical papilloma) were included (all women; mean age, 60.7 years). A multislit photon-counting tomosynthesis system was utilized (spectral imaging) to produce both low- and high-energy tomographic data (below and above the k edge of iodine, respectively) in a single scan, which allowed for dual-energy visualization of iodine. Images were obtained prior to contrast material administration and 120 and 480 seconds after contrast material administration. Four readers independently assessed the images along with conventional mammograms, ultrasonographic images, and magnetic resonance images. Glandular dose was estimated. RESULTS: Contrast agent was visible in the phantom model with simulated spherical tumor diameters as small as 5 mm. The average glandular dose was measured as 0.42 mGy per complete spectral imaging tomosynthesis scan of one breast. Because there were three time points (prior to contrast medium administration and 120 and 480 seconds after contrast medium administration), this resulted in a total dose of 1.26 mGy for the whole procedure in the breast with the abnormality. Seven of 10 cases were categorized as Breast Imaging Reporting and Data System score of 4 or higher by all four readers when reviewing spectral images in combination with mammograms. One lesion near the chest wall was not captured on the spectral image because of a positioning problem. CONCLUSION: The use of contrast-enhanced tomographic mammography has been demonstrated successfully in patients with promising diagnostic benefit. Further studies are necessary to fully assess diagnostic sensitivity and specificity.

Evaluation of contrast-enhanced digital mammography.

PURPOSE: The goal of this prospective study was to evaluate the possible diagnostic benefits of contrast-enhanced digital mammography (CEDM) over conventional mammography. MATERIALS AND METHODS: Our analysis included data from 70 patients with a total of 80 lesions (30 malignant and 50 benign). A series of contrast-enhanced images was acquired from each patient using a modified imaging system (GE Senographe 2000D with copper filter) suitable for displaying iodine contrast medium. After the mask image had been taken, the contrast medium was administered using a dosage of 1ml/kg body weight at a rate of 4ml/s. Three contrast-enhanced images in the cranio-caudal projection plane were then captured at intervals of 60s. The mask image was logarithmically subtracted from the contrast-enhanced images. We performed a ROC analysis of diagnostic quality with three readers. RESULTS: On average, 5.66 more malignant lesions were detected with the addition of digital dynamic contrast mammography versus conventional mammography alone. The sensitivity was increased from an average of 0.43 in conventional mammography to an average of 0.62 with contrast mammography. Even in dense breast parenchyma, the sensitivity increased from an average of 0.35-0.59. In the multi-reader-ROC analyses of all readers, the differences in the AUC with p=0.02 (BI-RADS) proved statistically significant in all cases. The Wilcoxon test showed that Readers I and II primarily used the CEDM to upgrade enhancing lesions to a higher BI-RADS category or a higher probability of malignancy. These two readers benefited most from the CEDM in the ROC analysis. CONCLUSION: Overall, we conclude that the addition of dynamic digital subtraction mammography to conventional mammography can significantly improve diagnostic quality. The increased sensitivity is particularly pronounced in the case of dense breast tissue.

Neuroimaging by 320-row CT: is there a diagnostic benefit or is it just another scanner? A retrospective evaluation of 60 consecutive acute neurological patients.

320-row CT enables dynamic CT angiography (4D CTA) of the entire intracranial circulation and whole-brain perfusion imaging (CTP). Sixty acute patients with neurological symptoms underwent various 320-row CT-specific protocols, including combined 4D CTA and CTP. Clinical and neuroradiological records were assessed for presumptive diagnoses, final diagnoses, supplementary and follow-up imaging studies. Additional diagnostic benefits delivered by 320-row CT were noted. Out of 60 procedures, 59 were accomplished successfully. Ischemia (n = 19), intracerebral hemorrhage (n = 7) and transient ischemic attacks (n = 10) were the major final diagnoses. Except one small cortical and two small subcortical infarctions all ischemias were diagnosed. All hemorrhages were diagnosed together with their underlying vascular pathology in five atypical cases. In conclusion, 320-row CT is a technically robust procedure being suitable for comprehensive neuroimaging of acute patients. It can provide dynamic angiographic and perfusion data of the whole brain and can deliver additional diagnostic information not available by standard CT.

4-D Imaging in cerebrovascular disorders by using 320-slice CT: feasibility and preliminary clinical experience.

RATIONALE AND OBJECTIVES: The authors report study protocols and initial clinical experience in assessing patients with acute and chronic cerebrovascular disorders using the recently introduced technique of volume computed tomography (VCT). MATERIALS AND METHODS: Thirteen patients with presumptive cerebrovascular insufficiency underwent VCT using a 320-slice scanner (detector width, 160 mm), including time-resolved whole-brain perfusion and cerebral angiography (four-dimensional computed tomographic angiography [CTA] and computed tomographic perfusion [CTP]). Unenhanced cranial CT (cCT) and helical cervicocranial CT (three-dimensional CTA) were added according to clinical requirements. Study protocols are presented, and image quality, data management, and radiation exposure issues are discussed. RESULTS: In 12 of 13 patients, the procedure was performed successfully on admission; in the other patient, the study was aborted for clinical reasons and repeated. Total scan time amounted to about 5 minutes, and data reconstruction times were up to 10 minutes. About 9000 primary images were generated, partially in the enhanced Digital Imaging and Communications in Medicine format, thus requiring new data postprocessing and management strategies. Image artifacts restricted the use of single-rotation cCT and incremental VCT (three-dimensional CTA). Overall exposure figures (computed tomographic dose index and dose-length product) were increased by 65% on average when three-dimensional CTA was added to volume cCT and four-dimensional CTA and CTP (5.0 mSv and 2178 mGy . cm, respectively). CONCLUSION: Preliminary clinical experience indicates that whole-brain four-dimensional CTA and CTP is a robust technique that provides relevant clinical information with respect to whole-brain perfusion as well as cerebral hemodynamics. The exposure benefit of deriving time-resolved perfusion and vessel images from one source data set is compromised when adding three-dimensional CTA to the protocol. Other acquisition techniques specific to VCT, such as single-rotation cCT and incremental three-dimensional CTA, suffer from restrictions in terms of image quality at present.

Thick slices from tomosynthesis data sets: phantom study for the evaluation of different algorithms.

Tomosynthesis is a 3-dimensional mammography technique that generates thin slices separated one to the other by typically 1 mm from source data sets. The relatively high image noise in these thin slices raises the value of 1-cm thick slices computed from the set of reconstructed slices for image interpretation. In an initial evaluation, we investigated the potential of different algorithms for generating thick slices from tomosynthesis source data (maximum intensity projection-MIP; average algorithm-AV, and image generation by means of a new algorithm, so-called softMip). The three postprocessing techniques were evaluated using a homogeneous phantom with one textured slab with a total thickness of about 5 cm in which two 0.5-cm-thick slabs contained objects to simulate microcalcifications, spiculated masses, and round masses. The phantom was examined by tomosynthesis (GE Healthcare). Microcalcifications were simulated by inclusion of calcium particles of four different sizes. The slabs containing the inclusions were examined in two different configurations: adjacent to each other and close to the detector and with the two slabs separated by two 1-cm thick breast equivalent material slabs. The reconstructed tomosynthesis slices were postprocessed using MIP, AV, and softMip to generate 1-cm thick slices with a lower noise level. The three postprocessing algorithms were assessed by calculating the resulting contrast versus background for the simulated microcalcifications and contrast-to-noise ratios (CNR) for the other objects. The CNRs of the simulated round and spiculated masses were most favorable for the thick slices generated with the average algorithm, followed by softMip and MIP. Contrast of the simulated microcalcifications was best for MIP, followed by softMip and average projections. Our results suggest that the additional generation of thick slices may improve the visualization of objects in tomosynthesis. This improvement differs from the different algorithms for microcalcifications, speculated objects, and round masses. SoftMip is a new approach combining features of MIP and average showing image properties in between MIP and AV.

Observer variability in screen-film mammography versus full-field digital mammography with soft-copy reading.

Full-field digital mammography (FFDM) with soft-copy reading is more complex than screen-film mammography (SFM) with hard-copy reading. The aim of this study was to compare inter- and intraobserver variability in SFM versus FFDM of paired mammograms from a breast cancer screening program. Six radiologists interpreted mammograms of 232 cases obtained with both techniques, including 46 cancers, 88 benign lesions, and 98 normals. Image interpretation included BI-RADS categories. A case consisted of standard two-view mammograms of one breast. Images were scored in two sessions separated by 5 weeks. Observer variability was substantial for SFM as well as for FFDM, but overall there was no significant difference between the observer variability at SFM and FFDM. Mean kappa values were lower, indicating less agreement, for microcalcifications compared with masses. The lower observer agreement for microcalcifications, and especially the low intraobserver concordance between the two imaging techniques for three readers, was noticeable. The level of observer agreement might be an indicator of radiologist performance and could confound studies designed to separate diagnostic differences between the two imaging techniques. The results of our study confirm the need for proper training for radiologists starting FFDM with soft-copy reading in breast cancer screening.

In patients with DCIS: is it sufficient to histologically examine only those tissue specimens that contain microcalcifications?

The purpose was to investigate in patients with histologically proven DCIS to what extent there is agreement between radiographically proven microcalcifications of specimens obtained by vacuum-assisted biopsy and the histologic diagnosis of microcalcifications and DCIS, and second, to assess the accuracy of biopsy in relation to the number of specimens obtained in patients with high-grade and low-grade DCIS. Four hundred twenty specimens from 35 patients who were diagnosed with DCIS were examined radiographically and histologically for the presence of microcalcifications. The results were analyzed using the McNemar-test. In addition, the average numbers of biopsy specimens necessary for diagnosing low-grade DCIS and high-grade DCIS were compared using the t-test. Specimen radiography had a PPV of 0.50 and a NPV of 0.85 for the demonstration of DCIS. Differences in localization between radiographically proven microcalcifications and DCIS were statistically significant (p<0.01). The difference between the mean numbers of specimens required per patient for correctly diagnosing high-grade or low-grade DCIS was statistically significant (p<0.01). Specimen radiography is very limited in identifying those specimens that are crucial for diagnosing DCIS. The rate of underestimation is expected to be higher for low-grade than for high-grade DCIS. The findings suggest that all samples obtained by vacuum-assisted breast biopsy should be histologically examined.

Evaluation of 11-gauge and 9-gauge vacuum-assisted breast biopsy systems in a breast parenchymal model.

RATIONALE AND OBJECTIVES: To compare three commercially available vacuum-assisted breast biopsy systems for tissue yield, length and fragmentation of specimens. MATERIALS AND METHODS: Specimens were acquired from radiolucent (bacon) and radioopaque (turkey breast) tissue using three different commercially available vacuum-assisted breast biopsy devices. Two systems (system 1 and 2) were equipped with 11 G needles, one system (system 3) with a 9 G needle. As for systems 1 and 2 a second chamber for applying the vacuum is attached to the needle, the external maximum diameter was identical for all three systems. 48 specimens were taken out for each tissue type and for each device. Specimens were measured for total weight, individual length, and number of fragments. Differences between groups were analyzed using analysis of variance (ANOVA) and Student's t-test. RESULTS: For both tissue types, system 1 and 2 showed similar results, for system 3 tissue weight and length of specimens were larger. Differences in lengths and weight were statistically significant between system 1 and 3 and system 2 and 3 (ANOVA, P < 0.05). Differences between length and weight were statistically significant between tissue 1 and 2 for all devices (t-Test < 0.05). CONCLUSION: As for system 3 a larger tissue yield was obtained with the same number of specimens compared to systems 1 and 2, it can be assumed that the same diagnostic accuracy as for systems 1 and 2 may be achieved for system 3 with less passes through tissue.

Contrast-to-noise ratios of different elements in digital mammography: evaluation of their potential as new contrast agents.

PURPOSE: To determine the contrast-to-noise ratios (CNRs) of different elements at different energies using various anode/filter combinations currently employed in digital mammography. The elements investigated included not only elements already used in conventional contrast agents such as gadolinium and iodine but also other elements to investigate their potential as mammographic contrast agents. MATERIALS AND METHODS: The CNRs of 20 mmol/L bismuth (Bi), gadolinium (Gd), ytterbium (Yb), dysprosium (Dy), and iodine (I) were determined at different slice thicknesses (0.25, 0.5, and 1 cm) of the element solution with an additional 4-cm Plexiglas in relation to water (to simulate dense glandular tissue), oil, and air. The following anode/filter combinations were used: Mo/Mo in the range of 22-34 kVp, Mo/Rh in the range of 36-40 kVp, Rh/Rh in the range of 42-46 kVp, and Mo/Cu in the range of 47-49 kVp. In the range of 22-46 kVp, the mAs were chosen to achieve a fairly uniform dose range (of 4.38-4.71 mGy). Doses were measured using the PTW DIADOS diagnostic dosimeter. The element solutions were examined with a GE Senographe 2000D. RESULTS: Bismuth showed the best CNR for all energies investigated and in relation to both water and oil. In the energy range below 46 kVp, bismuth (CNR at 30 kVp/50 mAs and 1/0.5/0.25 cm slice thickness: 9.9/6.1/3.4) was followed by Yb (5.9/3.5/2.0), Dy (5.3/3.2/1.9), Gd (4.2/2.5/1.6), and iodine (2.4/1.8/1.5). Bismuth had the best CNR relative to both water (values given above) and oil (Bi: 20.7/11.2/5; Yb: 16.9/8.6/3.6; Dy: 16.6/8.4/3.5; Gd: 15.21/7.5/3.2; I: 13.8/6.3/3.2). The CNR of Bi was also superior to that of the other elements investigated at high energy in combination with copper filters (eg, CNR at 49 kVp Mo/Cu at slice thicknesses of 1/0.5/0.25 cm, relative to water: 9.6/6.0/4.0) but now followed by iodine (7.9/5.3/3.5), Yb (5.8/4.0/2.9), Dy (5.4/3.7/2.8), and Gd (4.7/3.2/2.7). Iodine was the only element of those investigated whose contrast-to-noise ratio was improved with the use of a copper filter at high energies based on its K-edge (increase in CNR from 2.9 to 7.9 from 40 to 49 kVp at 1-cm slice thickness). Nevertheless, the improved CNR of iodine was below that of Bi at low energies and for Mo/Mo or Mo/Rh filters. The contrast of water/fat tended to decrease slightly at higher energies (CNR of water/air at 42 kVp: 33.9, at 48 kVp: 25.6; CNR of oil/air at 42 kVp: 23.8, at 48 kVp: 21.9). CONCLUSION: Copper filters and higher energies are useful for visualizing iodine-based contrast agents in contrast-enhanced mammography because they markedly improve the CNR relative to water. This technique further benefits from the fact that the CNR of water and fat relative to air markedly decreases at higher energies and with the use of copper filters. Bismuth was found to have a much better CNR than iodine for all energies investigated including the low energy ranges typically used in mammography. These results suggest that bismuth is a potential candidate for a specific mammographic contrast agent.

Breast lesion detection and classification: comparison of screen-film mammography and full-field digital mammography with soft-copy reading--observer performance study.

PURPOSE: To retrospectively compare screen-film and full-field digital mammography with soft-copy interpretation for reader performance in detection and classification of breast lesions in women in a screening program. MATERIALS AND METHODS: Regional ethics committee approved the study; signed patient consents were obtained. Two-view mammograms were obtained with digital and screen-film systems at previous screening studies. Six readers interpreted images. Interpretation included Breast Imaging Reporting and Data System (BI-RADS) and five-level probability-of-malignancy scores. A case was one breast, with two standard views acquired with both screen-film mammography and digital mammography. The standard for an examination with normal findings was classification of normal (category 1) assigned by two independent readers; for cases with benign findings, the standard was benign results at diagnostic work-up in patients who were recalled. Cases with normal or benign findings that manifested as neither interval cancer nor as cancer at subsequent screening were considered the standard. All cancers were confirmed histologically. Images were interpreted by readers in two sessions 5 weeks apart; the same case was not seen twice in any session. Receiver operating characteristic (ROC) analysis and, for a given true-positive fraction, 2 x 2 table analysis and the McNemar test were used. For binary outcome, classification of BI-RADS category 3 or higher was defined as positive for cancer. RESULTS: Cases with proved findings (n = 232) were displayed: 46 with cancers, 88 with benign findings, and 98 with normal findings. ROC analysis for all readers and all cases revealed a higher area under ROC curve (A(z)) for digital mammography (0.916) than for screen-film mammography (0.887) (P = .22). Five of six readers had a higher performance rating with digital mammography; one of five demonstrated a significant difference in favor of digital mammography with A(z) values; two showed a significant difference in favor of digital mammography with ROC analysis for a given false-positive fraction (P = .01 and .03, respectively). For cases with cancer, digital mammography resulted in correct classification of an average of three additional cancers per reader. For digital versus screen-film mammography, 2 x 2 table analysis for cancers revealed a higher true-positive rate; for benign masses, a higher true-negative rate. Neither of these differences nor any others from analysis of subgroups between the modalities were significant. CONCLUSION: Digital mammography allowed correct classification of more breast cancers than did screen-film mammography. A(z) value was higher for digital mammography; this difference was not significant.

Digital mammography using iodine-based contrast media: initial clinical experience with dynamic contrast medium enhancement.

PURPOSE: We sought to evaluate the potential of dynamic contrast enhancement after the intravenous administration of an iodine-based contrast medium in full-field digital mammography. MATERIAL AND METHODS: A protocol for image acquisition was established for contrast-enhanced mammography and the mammography unit (Senographe 2000D, GE Healthcare, Buc, France) changed as required. The effect of the protocol parameters on imaging was investigated. Subsequently, 21 patients with 25 suspicious lesions of the breast (10 benign, 1 borderline, and 14 malignant) underwent mammography with administration of an iodine-based contrast medium (Ultravist 370, Schering AG, Berlin, Germany), after approval of ethical committee as well as permission of German federal office for Radiation protection, and informed consent from each patient was obtained. Three sequential digital mammographic images of the respective breast were acquired after administration of the contrast medium at a dose of 1 mL/kg body weight and a flow of 4 mL/s. The postcontrast images were acquired 60, 120, and 180 seconds after administration. Subsequently, the precontrast image was logarithmically subtracted from the postcontrast images. Enhancement of the lesions was measured in absolute terms as well as relative to the enhancement of the glandular tissue. The subtracted images were evaluated for lesion depiction and dynamic contrast enhancement. Lesion-enhancement kinetics were compared with the histologic findings. RESULTS: All malignant lesions were identified on the contrast-enhanced images of digital mammography. Three of the tumors (2 malignant, 1 benign) were detected only by contrast-enhanced mammography and not by standard mammography. Dynamic enhancement curves of benign and malignant tumors in contrast-enhanced mammography look similar to the curves known from gadolinium-enhanced magnetic resonance imaging. Nevertheless differentiation between malignant and benign tumors based on the enhancement patterns cannot be directly taken over from magnetic resonance imaging, as suggested by our initial results. The results are somewhat better when tumor enhancement relative to surrounding glandular tissue is used instead of absolute enhancement. CONCLUSION: The results of this preliminary study suggest that contrast-enhanced digital mammography is a potentially useful tool for the detection and the differentiation of benign and malignant breast lesions.

New contrast media designed for x-ray energy subtraction imaging in digital mammography.

RATIONALE AND OBJECTIVES: In contrast-enhanced dual-energy subtraction imaging 2 images acquired postcontrast media administration at different energies are subtracted to highlight structures hidden in the absence of contrast media. X-ray spectra of the newly developed digital full-field mammography units (GE Senographe 2000 D) are dominated by the emission lines of the Mo or Rh anodes. The K-edge of Zirconium (Zr) is flanked by these 2 emission lines. Thus, the attenuation of Zr should experience a pronounced change of attenuation in parallel with a change of anodes. Under clinically relevant conditions, the contrasting behavior of Zr should be compared with that of other elements having K-edge energies outside the window spanned by the 2 anode emission lines. METHODS: Solutions containing the contrasting elements Br, Y, Zr, I, and Gd were investigated for dual-energy subtraction in digital mammography with the 2 anode/filter settings (Mo/Mo and Rh/Rh). These solutions were investigated in phantom studies in the energy range conventionally used in mammography. Additionally, the contrasting behavior of Zr and I was compared in an in vivo study in rats. RESULTS: The sweeping over the K-edge by alternating between the Mo and Rh anodes increases the detection of Zr in energy subtraction imaging at constant high voltage. This procedure does not lead to sufficient contrast enhancement for iodine-based contrast media which become detectable by increasing the high voltage to 40-49 kV. CONCLUSION: The instrumental and physical data outlined predestine Zr as contrasting element with a high potential for energy subtraction imaging in digital mammography in the energy range conventionally applied.