Dual-energy Chest Radiography Research Articles

Framework for dual-energy-like chest radiography image synthesis from single-energy computed tomography based on cycle-consistent generative adversarial network.

Dual-energy (DE) chest radiography (CXR) enables the selective imaging of two relevant materials, namely, soft tissue and bone structures, to better characterize various chest pathologies (i.e., lung nodule, bony lesions, etc.) and potentially improve CXR-based diagnosis. Recently, deep-learning-based image synthesis techniques have attracted considerable attention as alternatives to existing DE methods (i.e., dual-exposure-based and sandwich-detector-based methods) because software-based bone-only and bone-suppression images in CXR could be useful. The objective of this study was to develop a new framework for DE-like CXR image synthesis from single-energy computed tomography (CT) based on a cycle-consistent generative adversarial network. The core techniques of the proposed framework are divided into three categories: (1) data configuration from the generation of pseudo CXR from single energy CT, (2) learning of the developed network architecture using pseudo CXR and pseudo-DE imaging using a single-energy CT, and (3) inference of the trained network on real single-energy CXR. We performed a visual inspection and comparative evaluation using various metrics and introduced a figure of image quality (FIQ) to consider the effects of our framework on the spatial resolution and noise in terms of a single index through various test cases. Our results indicate that the proposed framework is effective and exhibits potential synthetic imaging ability for two relevant materials: soft tissue and bone structures. Its effectiveness was validated, and its ability to overcome the limitations associated with DE imaging techniques (e.g., increase in exposure dose owing to the requirement of two acquisitions, and emphasis on noise characteristics) via an artificial intelligence technique was presented. The developed framework addresses X-ray dose issues in the field of radiation imaging and enables pseudo-DE imaging with single exposure.

Artifacts in Dual-Energy Chest Radiography

The aim of this research is to analyze artifacts in dual-energy chest radiography: their appearance and how to eliminate them. A total of 252 patients were examined in random order by using dual-energy radiography. As a result, 756 images were analyzed. This review describes the most important and most prevalent of them, including motion artifacts - 58 (41,4%), quantum noise – 55 (39,2%) and decreasing the suppression of bony elements on soft-tissue-selective images – 27 (19,3%). The main reason of dual-energy chest radiography artifacts is incorrect image reconstruction parameters.

Estimating the accuracy of dual energy chest radiography for coronary calcium detection with lateral or anteroposterior orientations.

Coronary artery calcium (CAC) scoring with CT has been studied as a risk stratification tool for cardiovascular disease. However, concerns remain from the radiation dose, economic expense, and incidental findings associated with this exam. Dual energy chest X-ray (DE CXR) has been proposed as an alternative, but validation of this technique remains limited. The purpose of this work was twofold: first, to estimate the sensitivity and specificity of DE CXR using simulation of patient datasets in a CAC screening cohort; second, to assess if sensitivity and specificity could be improved using a lateral instead of an anteroposterior (AP) orientation. We started from a cohort of 73 CAC scoring CT exams after exclusions for metal wires, data truncation, or with age outside 40-75 years. The fraction of CT CAC scores in the validation set of 0, 1-99, 100-299, and 300+ were 36, 25, 14, and 26%, respectively. CT datasets were decomposed on a voxel-by-voxel basis into mixtures of water and calcium according to CT number. DE CXR images were simulated using polyenergetic forward projection with scatter estimated from Monte Carlo. We assumed a technique of 60 and 120kVp for the dual energy acquisition. The tube current was scaled such that the estimated radiation dose from DE CXR was 10 times less than CAC scoring CT. Patient motion was not simulated. Two readers read the validation set in a blinded, randomized fashion, and estimated the amount of CAC in each DE CXR image using a semiquantitative 4-point scale. Although patients present on a spectrum of CAC severity, in the primary analysis, sensitivity and specificity were calculated by dichotomizing patients into two categories of CT CAC (Agatston) scores of either 0-99 or 100+. From the lateral orientation, average sensitivity between two readers was 69% (range, 69-69%), specificity was 85% (range, 84-86%), and area under the curve (AUC) was 0.81 (range, 0.80-0.81). From the AP orientation, average sensitivity was 35% (range, 31-38%), average specificity was 70% (range, 66-73%), and AUC was 0.54 (range, 0.53-0.55). Reader DE CXR scores agreed within 1 point of the 4-point scale on 97% of ratings from the lateral orientation and 80% from the AP orientation. From the lateral orientation, AUC increased when considering higher CT CAC score thresholds as disease positive; for thresholds of 1+, 300+, and 1000+, average AUC was 0.72, 0.81, and 0.92, respectively. From the AP orientation, AUC was 0.57, 0.55, and 0.61, respectively. DE CXR for CAC scoring may have higher diagnostic accuracy when acquired from the lateral orientation. The sensitivity and specificity of lateral DE CXR, when combined with its modest cost and radiation dose, suggest a possible role for this technique in screening coronary calcium in lower risk individuals. These estimates of diagnostic accuracy are derived from simulation of patient datasets and have not been corroborated with experimental or clinical images.

Effects of the technique parameters on the imaging performance of the dual-energy chest radiography

Conventional chest-radiography images are the most basic X-ray imaging system for diagnosing lung diseases. However, space-averaging increases due to the overlapping of anatomy structures as three-dimensional information are reduced in two-dimensions. A dual-energy X-ray imaging (DEI) technique for resolve space-averaging is used to increase the conspicuity of the lesion (i.e., lung nodule and calcification) by suppressing anatomical background noise. DEI techniques perform the weighted log-subtraction using the attenuation characteristics of the material for low- and high-kilovoltage-peak (kVp) X-rays. And the weighting factor is calculated to suppressing anatomical background. The objective of this research is to increase the lesion conspicuity in lung regions by optimizing DE technique parameters. DE technique has been applied to three phantoms mimicking the average, thick, and thicker adult chest thicknesses while maintaining the total incident dose equivalent to that used to obtain a single chest radiograph. Investigating DE technique parameters include the added filtration, the low- and high-kVp pair, and the dose allocation between low- and high-kVp setups at the total dose equal to or less than that of the conventional chest radiograph. And the DE image performances investigated through contrast-to-noise ratio, modulation-transfer function, noise-power spectrum, and detective quantum efficiency.

Development of a deep neural network for generating synthetic dual-energy chest x-ray images with single x-ray exposure

Dual-energy chest radiography (DECR) is a medical imaging technology that can improve diagnostic accuracy. This technique can decompose single-energy chest radiography (SECR) images into separate bone- and soft tissue-only images. This can, however, double the radiation exposure to the patient. To address this limitation, we developed an algorithm for the synthesis of DECR from a SECR through deep learning.To predict high resolution images, we developed a novel deep learning architecture by modifying a conventional U-net to take advantage of the high frequency-dominant information that propagates from the encoding part to the decoding part. In addition, we used the anticorrelated relationship (ACR) of DECR for improving the quality of the predicted images. For training data, 300 pairs of SECR and their corresponding DECR images were used. To test the trained model, 50 DECR images from Yonsei University Severance Hospital and 662 publicly accessible SECRs were used. To evaluate the performance of the proposed method, we compared DECR and predicted images using a structural similarity approach (SSIM). In addition, we quantitatively evaluated image quality calculating the modulation transfer function and coefficient of variation.The proposed model selectively predicted the bone- and soft tissue-only CR images from an SECR image. The strategy for improving the spatial resolution by ACR was effective. Quantitative evaluation showed that the proposed method with ACR showed relatively high SSIM (over 0.85). In addition, predicted images with the proposed ACR model achieved better image quality measures than those of U-net.In conclusion, the proposed method can obtain high-quality bone- and soft tissue-only CR images without the need for additional hardware for double x-ray exposures in clinical practice.

Enhanced coronary calcium visualization and detection from dual energy chest x-rays with sliding organ registration

We have developed a technique to image coronary calcium, an excellent biomarker for atherosclerotic disease, using low cost, low radiation dual energy (DE) chest radiography, with potential for widespread screening from an already ordered exam. Our dual energy coronary calcium (DECC) processing method included automatic heart silhouette segmentation, sliding organ registration and scatter removal to create a bone-image-like, coronary calcium image with significant reduction in motion artifacts and improved calcium conspicuity compared to standard, clinically available DE processing. Experiments with a physical dynamic cardiac phantom showed that DECC processing reduced 73% of misregistration error caused by cardiac motion over a wide range of heart rates and x-ray radiation exposures. Using the functional measurement test (FMT), we determined significant image quality improvement in clinical images with DECC processing (p < 0.0001), where DECC images were chosen best in 94% of human readings. Comparing DECC images to registered and projected CT calcium images, we found good correspondence between the size and location of calcification signals. In a very preliminary coronary calcium ROC study, we used CT Agatston calcium score >50 as the gold standard for an actual positive test result. AUC performance was significantly improved from 0.73 ± 0.14 with standard DE to 0.87 ± 0.10 with DECC (p = 0.0095) for this limited set of surgical patient data biased towards heavy calcifications. The proposed DECC processing shows good potential for coronary calcium detection in DE chest radiography, giving impetus for a larger clinical evaluation.

Evaluation of a commercial cardiac motion phantom for dual‐energy chest radiography

Misregistration due to cardiac motion causes artifacts in two‐exposure dual‐energy subtraction images, in both the soft‐tissue‐only image and the bone‐only image. Two previous investigations have attempted to avoid misregistration artifacts by using cardiac gating of the first and second exposures. The severity of misregistration was affected by the heart rate, the time interval between the low‐ and high‐energy exposures, the total duration of the two exposures, and the phase of the cardiac cycle at the start of the exposure sequence. We sought to determine whether a commercial phantom with a simulated beating heart can be use to investigate the factors affecting misregistration in dual‐energy chest radiography. We made dual‐energy images of the phantom in postero–anterior orientation using the indirect digital radiography system (GE XQ/i). We acquired digital images at heart rates between 40 beats per minute and 120 beats per minute and transferred them to a computer, where the area of the artifact on the silhouette of the heart was measured from both soft‐tissue‐only and bone‐only images. For comparison, we measured misregistration in clinical dual‐energy subtraction images by the same method. Generally speaking, without synchronization of the exposure sequence with the cardiac cycle, the area of the misregistration artifact increased with heart rate for both the phantom and clinical images. However, the phantom exaggerated the magnitude of misregistration relative to clinical images. Although this phantom was designed for horizontal operation and computed tomography imaging, it can be use in an upright configuration to simulate heart motion for investigation of dual‐energy misregistration artifacts and control.PACS numbers: 87.59.bf, 87.57.cf, 87.57N

Dynamic dual-energy chest radiography: a potential tool for lung tissue motion monitoring and kinetic study

Dual-energy chest radiography has the potential to provide better diagnosis of lung disease by removing the bone signal from the image. Dynamic dual-energy radiography is now possible with the introduction of digital flat-panel detectors. The purpose of this study is to evaluate the feasibility of using dynamic dual-energy chest radiography for functional lung imaging and tumor motion assessment. The dual-energy system used in this study can acquire up to 15 frames of dual-energy images per second. A swine animal model was mechanically ventilated and imaged using the dual-energy system. Sequences of soft-tissue images were obtained using dual-energy subtraction. Time subtracted soft-tissue images were shown to be able to provide information on regional ventilation. Motion tracking of a lung anatomic feature (a branch of pulmonary artery) was performed based on an image cross-correlation algorithm. The tracking precision was found to be better than 1 mm. An adaptive correlation model was established between the above tracked motion and an external surrogate signal (temperature within the tracheal tube). This model is used to predict lung feature motion using the continuous surrogate signal and low frame rate dual-energy images (0.1–3.0 frames per second). The average RMS error of the prediction was (1.1 ± 0.3) mm. The dynamic dual energy was shown to be potentially useful for lung functional imaging such as regional ventilation and kinetic studies. It can also be used for lung tumor motion assessment and prediction during radiation therapy.

CsI-detector-based dual-exposure dual energy in chest radiography for lung nodule detection: results of an international multicenter trial

To assess both sensitivity and specificity of digital chest radiography alone and in conjunction with dual-exposure dual-energy chest radiography for the detection and classification of pulmonary nodules. One hundred patients with a total of 149 lung nodules (3-45 mm; median, 11 mm) confirmed by CT were included in this study. Dual-exposure dual-energy chest radiographies of each patient were obtained using a CsI detector system. Experienced board-certified chest radiologists from four different medical centers in Europe reviewed standard chest radiographs alone and in conjunction with dual-energy images blinded and in random order. The reviewers rated the probability of presence, calcification and malignancy of all lung nodules on a five-point rating scale. Lesions detected were identified by applying a specific coordinate system to enable precise verification by the study leader. A receiver-operating characteristic (ROC) analysis was performed. In addition to the 149 true-positive CT proven lesions, 236 false-positive lung nodules were described in digital chest radiographies in conjunction with dual-energy chest radiographies. The cumulative sensitivity of chest radiography in conjunction with dual energy was 43%, specificity was 55%. For digital radiography alone, sensitivity was 35% and specifity was 83%. For the dual energy system, positive predictive value was 58%, and negative predictive value was 66% compared to the digital radiography with a positive predictive value of 59% and a negative predictive value of 65%. Areas under the curve in a ROC analysis resulted in 0.631 (95% confidence interval =0.61 to 0.65) for radiography with dual energy and 0.602 (95% confidence interval =0.58 to 0.63) for digital radiography alone. This difference was not statistically significant. For the detection of lesion calcification or the determination of malignancy, ROC analysis also failed to show significant differences. CsI-based flat-panel dual-exposure dual-energy imaging added to standard chest radiography did not show statistically significant improvement for the detection of pulmonary nodules, nor the identification of calcifications, nor the determination of malignancy.

Dual-energy chest radiography with a flat-panel digital detector: revealing calcified chest abnormalities.

The aim of this study was to assess the value of dual-energy chest radiography obtained using a cesium iodide flat-panel detector in addition to standard posteroanterior chest radiography for the detection of calcified chest abnormalities. The study included 20 patients with a total of 37 calcified chest lesions (16 pulmonary nodules, 17 mediastinal calcifications, and four pleural calcifications) as confirmed on CT. Twenty-eight locations in the chests of the same patients who were free of lesions were used as negative controls. Four radiologists reviewed posteroanterior chest radiographs in a blinded manner alone and in conjunction with dual-energy soft-tissue and bone images. We calculated sensitivity, specificity, the negative predictive value (NPV), and the positive predictive value (PPV) for lesion prediction. The Wilcoxon's and the Brunner and Langer's tests were performed for statistical analysis. For posteroanterior chest radiography, sensitivity was 36%, the PPV was 64%, and the NPV was 47%. When dual-energy images were added, sensitivity increased significantly to 66% (p < 0.05), the PPV to 76%, and the NPV to 62%. The specificity remained constant at 73%. Brunner and Langer's test revealed a highly significant difference between posteroanterior chest radiography and dual-energy imaging in the detection of calcified chest abnormalities (p < 0.01). Dual-energy images added to standard posteroanterior chest radiographs significantly improve the detection of calcified chest lesions.

Application of AMBER in single- and dual-energy digital imaging: improvement in noise level and display dynamic range.

The combination of computed radiography (CR) and advanced multiple beam equalization radiography (AMBER) was evaluated for both single- and dual-energy chest radiography. The improved signal-to-noise ratio found with CR and AMBER resulted in a better visualization of structures in the mediastinum and basal lung than that found with CR alone. For the central lung, no improvement was seen. Because of the compressed dynamic range with CR and AMBER, contrast on hard copies and video monitors could be high without a sacrifice in image latitude. Dual-energy images showed a considerably lower noise level. The combined use of AMBER and CR promises to overcome the dynamic range limitations of digital displays while improving signal-to-noise ratio.

Scanning slit chest radiography: a practical and efficient scatter control design.

To design a chest radiography unit that eliminates scatter. A scanning slit radiography unit that also incorporates a conventional grid was designed. The scanning slit geometry defined a 10-mm-wide beam at the image receptor. With a typical heavy-duty radiographic tube, the proposed system was capable of delivering an entrance skin exposure of up to 32 microC/kg for a 1-second scan. Monte Carlo simulations indicated that the scatter-to-primary ratio (S/P) imaged with the scanning slit system alone was 0.125 for a 35 x 43-cm, 23-cm-thick Lucite phantom imaged at 140 kVp. When the system was coupled with 4:1, 8:1, and 12:1 grids, S/P decreased to 0.063, 0.055, and 0.054, respectively. A scanning slit system coupled with a conventional grid is a practical and highly efficient method of controlling scatter and will result in minimal soft-tissue and bone measurement errors in dual-energy chest radiography.

Conventional chest radiography vs dual-energy computed radiography in the detection and characterization of pulmonary nodules.

We evaluated a single-exposure, phosphor-plate, dual-energy imaging device that produces, in addition to conventional chest radiographs, both tissue- and bone-selective images. Our purpose was to determine whether dual-energy radiography was more accurate than routine chest radiography for detection and characterization of pulmonary nodules. Two hundred patients undergoing chest CT were asked to volunteer to have dual-energy and conventional chest radiographs obtained immediately before or after their CT scan. Radiographs from a subset of 50 of these patients with 116 CT-detected nodules and 10 patients with normal findings on CT scans of the chest were presented to the observers for the nodule detection study. Similarly, radiographs from a subset of 29 patients with 20 calcified and 20 uncalcified nodules were presented to five observers to determine nodule calcification. Dual-energy images were produced by filtering the X-ray tube output with a gadolinium sheet while using a multiple phosphor plate receptor. A dual-energy triad of images consisting of a conventional image, a tissue-selective image, and a bone-selective image were produced. The conventional chest radiographs and dual-energy image sets were presented to observers in random order. Data from a free response receiver operating curve and a receiver operating curve were generated for nodule detection and characterization, respectively. By using the dual-energy images, all five observers improved their ability to diagnose pulmonary nodules (p = .0005) and to characterize nodules as calcified (p = .005). By eliminating rib shadows with tissue-selective images and enhancing calcified structures with bone-selective images, dual-energy chest radiography improved the ability of all observers, regardless of expertise, to detect and characterize pulmonary nodules.

Mineral content of bone: measurement by energy subtraction digital chest radiography.

Several methods are presently available for measuring the mineral content of bone. Those in widespread use include dual-photon absorptiometry and quantitative CT. The feasibility of using dual-energy digital chest radiography for determination of the mineral content of posterior ribs on digital chest images was studied by using a prototype unit. The results showed a significant difference in the mineral density of the posterior ribs of control subjects and those of patients who had osteoporosis (251.1 +/- 36 mg Ca2+/cm2 of rib vs 158.8 +/- 48 mg Ca2+/cm2, p less than or equal to .01) and a close correlation with values obtained by dual-photon absorptiometry of the lumbar spine (r = .77). The results suggest that this technique can provide an accurate assessment of the presence or absence of osteoporosis.