Original ResearchFree Access

Qualitative and Quantitative Assessment of Emphysema Using Dark-Field Chest Radiography

Published Online:https://doi.org/10.1148/radiol.212025

Abstract

Background

Dark-field chest radiography allows for assessment of lung alveolar structure by exploiting wave optical properties of x-rays.

Purpose

To evaluate the qualitative and quantitative features of dark-field chest radiography in participants with pulmonary emphysema as compared with those in healthy control subjects.

Materials and Methods

In this prospective study conducted from October 2018 to October 2020, participants aged at least 18 years who underwent clinically indicated chest CT were screened for participation. Inclusion criteria were an ability to consent to the procedure and stand upright without help. Exclusion criteria were pregnancy, serious medical conditions, and any lung condition besides emphysema that was visible on CT images. Participants were examined with a clinical dark-field chest radiography prototype that simultaneously acquired both attenuation-based radiographs and dark-field chest radiographs. Dark-field coefficients were tested for correlation with each participant’s CT-based emphysema index using the Spearman correlation test. Dark-field coefficients of adjacent groups in the semiquantitative Fleischner Society emphysema grading system were compared using a Wilcoxon Mann-Whitney U test. The capability of the dark-field coefficient to enable detection of emphysema was evaluated with receiver operating characteristics curve analysis.

Results

A total of 83 participants (mean age, 65 years ± 12 [standard deviation]; 52 men) were studied. When compared with images from healthy participants, dark-field chest radiographs in participants with emphysema had a lower and inhomogeneous dark-field signal intensity. The locations of focal signal intensity loss on dark-field images corresponded well with emphysematous areas found on CT images. The dark-field coefficient was negatively correlated with the quantitative CT-based emphysema index (r = –0.54, P < .001). Participants with Fleischner Society grades of mild, moderate, confluent, or advanced destructive emphysema exhibited a lower dark-field coefficient than those without emphysema (eg, 1.3 m–1 ± 0.6 for participants with confluent or advanced destructive emphysema vs 2.6 m–1 ± 0.4 for participants without emphysema; P < .001). The area under the receiver operating characteristic curve for detection of mild emphysema was 0.79.

Conclusion

Pulmonary emphysema leads to reduced signal intensity on dark-field chest radiographs, showing the technique has potential as a diagnostic tool in the assessment of lung diseases.

© RSNA, 2022

See also the editorial by Hatabu and Madore in this issue.

Summary

Dark-field chest radiography shows dark-field signal intensity losses that correspond to emphysematous areas found on chest CT images in participants with emphysema.

Key Results

  • ■ In a prospective study of 83 individuals, focal loss of signal intensity on dark-field chest radiographs was qualitatively associated with regions of focal emphysema on CT images.

  • ■ The dark-field coefficient was negatively correlated with the quantitative CT-based emphysema index (r = –0.54; P < .001).

  • ■ The dark-field coefficient was lower in participants with mild (mean, 2.2 m–1 ± 0.4 [standard deviation]; P = .02), moderate (2.1 m–1 ± 0.7; P = .01), or confluent or advanced destructive (1.3 m–1 ± 0.6; P < .001) emphysema than in control subjects (2.6 m–1 ± 0.4).

Introduction

Chronic obstructive pulmonary disease (COPD) is a major contributor to morbidity and mortality worldwide, with smoking and air pollution being the main risk factors. Its key components are chronic bronchitis and pulmonary emphysema (1). Early diagnosis of this disease is crucial for treatment and smoking cessation programs. However, current widely applied diagnostic tests do not work well for the early stages of this disease: Spirometry shows low sensitivity (2) and strongly depends on patient cooperation (3). Chest radiography lacks sensitivity and is not recommended for COPD diagnosis (4). Chest CT provides three-dimensional information and enables emphysema detection through classification of lung voxels by attenuation (in Hounsfield units) (5,6). An emphysematous voxel of lung tissue is defined as having an attenuation of less than –950 HU at full inspiration. The resulting fraction of emphysematous lung tissue constitutes the emphysema index (EI). However, attenuation measurements of emphysema are known to vary with patient dose, section thickness, hardware, and reconstruction algorithm (7). Therefore, only an overall EI of 6% or more is considered significant by the Fleischner Society (8). Additionally, CT is more expensive and has a higher radiation dose (standard chest CT, 8 mSv; low-dose chest CT, 1–2 mSv; ultra-low-dose chest CT, 0.07 mSv) (9,10), making it less suitable for emphysema screening.

X-ray dark-field imaging was introduced in 2008 (11). It is particularly sensitive to microstructures, as the image contrast is formed by multiple refractions of the x-ray beam at material interfaces (small-angle scattering), such as the alveoli in the lung (11,12). Various studies have shown that dark-field imaging is particularly useful for lung imaging and that an impairment of the alveolar structure leads to a reduction of dark-field signal. This effect has been shown in animal models for fibrosis (13), neonatal lung injury (14), lung cancer (15), and the detection and staging of pulmonary emphysema (16,17). For pulmonary emphysema in particular, dark-field chest radiography showed a strong signal decrease in affected lung areas (17).

The feasibility of dark-field x-ray imaging has been demonstrated in animal (18,19) and ex vivo human studies (20,21). Recently, a clinical prototype system was constructed (22), and a clinical study on the diagnostic value of dark-field chest radiography is ongoing. Building on the recent evaluation of dark-field chest radiography images in healthy individuals (23), the purpose of the present cross-sectional study was to evaluate the qualitative and quantitative features of x-ray dark-field images in participants with pulmonary emphysema.

Materials and Methods

Participants

This prospective study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). Approval of the institutional review board and the national radiation protection agency was obtained prior to this study (Ethics Commission of the Medical Faculty, Technical University of Munich, Germany; reference no. 166/20S). Participants gave their written informed consent. Between October 2018 and October 2020, participants aged at least 18 years who underwent chest CT as part of their diagnostic work-up were screened for study participation. Inclusion criteria were the ability to consent and the ability to stand upright without help. Exclusion criteria were pregnancy, lung cancer, pleural effusion, atelectasis, air-space disease, ground-glass opacities, and pneumothorax. Figure 1 shows the study flow and selection process. A total of 40 participants were previously described in a study by Gassert et al (23) evaluating the characteristics of dark-field chest radiographs in healthy humans. A total of 77 participants were also studied by Willer et al (22), who assessed the diagnostic accuracy of dark-field imaging in emphysema diagnosis based on a reader study of dark-field chest radiography findings and functional lung testing without analyzing the quantitative dark-field signal.

Flowchart shows participant selection criteria. Between October 2018                         and October 2020, 83 participants were included in the study.

Figure 1: Flowchart shows participant selection criteria. Between October 2018 and October 2020, 83 participants were included in the study.

Prototype Dark-Field Chest X-Ray System

The basic principle of grating-based x-ray dark-field imaging was introduced in 2008 (11). The schematic of the setup used in this work and further technical details have been described by Gassert et al (23). The system consists of a conventional imaging system with a diagnostic x-ray tube (MRC 200 0508 ROT-GS 1003; Philips Medical Systems) operated at 70 kVp and a flat-panel detector (PIXIUM 4343 F4; Trixell) in combination with a three-grating interferometer (G0, G1, G2) enabling the detection of small-angle scattering. The second grating (G1) is the phase grating, which imparts a periodic intensity modulation on the beam that can be altered by a sample in the beam path. Attenuation of the x-rays leads to a reduction of the mean, while small-angle scattering reduces the amplitude of the intensity modulation. Since this modulation is much smaller than the pixel size of typical detectors, a third grating (G2) is used to analyze this pattern by taking multiple exposures with different relative grating positions. For conventional x-ray sources with low coherence, an additional first source grating (G0) is necessary (24). Both an attenuation-based image similar to a conventional radiograph and a dark-field image are calculated from the acquired raw images of one acquisition.

All participants were examined with one acquisition each in posteroanterior and lateral orientations at full inspiration. The acquisition time was about 7 seconds. The effective dose (participant collective median) was 37 µSv for the posteroanterior images (25) and 46 µSv for the lateral images.

Quantitative Dark-Field Imaging

For all dark-field images shown in this work, the same window level and window width were used, where black corresponds to 0 HU or less and white corresponds to a dark-field signal intensity of 0.6 or more. For the quantitative comparison, every subject’s average dark-field coefficient was calculated according to Gassert et al (23). It is the integral of the dark-field signal over the whole lung area, manually selected from the images to cover the whole lung, divided by the lung volume of the subject, determined from the posteroanterior and lateral attenuation images, both acquired with the dark-field chest radiography system, following Pierce et al (26). This average dark-field coefficient corresponds to the mean dark-field signal generated by the subject’s lung tissue per unit path length that the x-rays pass through. This quantity is proportional to the alveolar density of the lung tissue under examination. Thus, more healthy lung tissue will cause higher dark field signal intensity, while less lung tissue (emphysema) will cause lower dark field signal intensity.

CT Emphysema Evaluation

CT was performed on one of two CT scanners (iCT and IQon Spectral CT; Philips) with the following parameters for iCT and IQon Spectral CT scanners, respectively, according to routine clinical protocols: collimation, 128 × 0.6 mm and 64 × 0.6 mm; pixel spacing, 0.4 and 0.3 mm; pitch factor, 0.8 and 0.9; peak tube voltage, 120 kVp; modulated tube current, 102–132 mA. Images were reconstructed at 0.9 mm section thickness with a lung-specific convolution kernel.

Semiquantitative assessment of emphysema was performed by three radiologists (J.H.B., A.P.S., A.A.F.; 4, 7, and 13 years of experience in CT imaging, with A.A.F. also having subspecialty experience in thoracic radiology) based on CT images. Participants were classified as nonemphysematous (emphysema was absent), or their emphysema was rated on a five-point Likert scale grading system (trace, mild, moderate, confluent, or advanced destructive emphysema) using the Fleischner Society guidelines for emphysema scoring (8).

Emphysema quantification (emphysema index [EI]), was performed using dedicated commercial software (IntelliSpace Portal, version 11.1.1; Philips). Its working principle has been previously described by Müller et al (5) and Wang et al (6): Essentially, the lung is segmented from the chest CT data, and a threshold of –950 HU is applied to distinguish between the healthy lung tissue and the emphysematous tissue (Fig 2). The EI is determined by the percentage of emphysematous (<–950 HU) lung voxels. According to the Fleischner Society (8), only an EI of 6% or more is considered significant. Below this value, no definite statement on emphysema presence can be made. Therefore, we assessed the correlation between dark-field coefficient and EI twice, once for all participants and once for only those participants with an EI of 6% or more.

Emphysema quantification on a CT scan in a 71-year-old man. (A)                         Example sections in axial and coronal reformation. All voxels within the                         lung with an attenuation less than –950 HU are labeled as                         emphysematous (red). (B) The three-dimensional emphysema map is projected                         along the sagittal axis to generate an overlay of the CT-based attenuation                         and the emphysema projection.

Figure 2: Emphysema quantification on a CT scan in a 71-year-old man. (A) Example sections in axial and coronal reformation. All voxels within the lung with an attenuation less than –950 HU are labeled as emphysematous (red). (B) The three-dimensional emphysema map is projected along the sagittal axis to generate an overlay of the CT-based attenuation and the emphysema projection.

For visual comparison of CT results with dark-field images, the three-dimensional CT emphysema map was projected along the sagittal axis to generate an overlay of the CT-based attenuation image and the projected emphysema quantification (Fig 2). We used a two-dimensional color map, relating the color of each pixel to its quantitative CT emphysema ratio and its brightness to the attenuation. The intensity of the red color increases with the number of emphysematous voxels relative to the total number of lung voxels along the posteroanterior projection plane of that pixel (a ray sum) and saturates to maximum red intensity at 50% or greater.

Statistical Analyses

Statistical analysis was performed with Python (version 3.6.9; https://www.python.org/) using the packages NumPy (version 1.19.2) and SciPy (version 1.5.2) with P < .05 chosen to indicate a significant difference. The participant parameters age, weight, and lung volume were tested for significant differences between participants with emphysema and control subjects with a Student t test. For the parameter sex, a χ2 test was used. The dark-field coefficient and the CT-based EI were analyzed for correlation for all participants and additionally for participants with an EI greater than or equal to 6% only using Spearman r correlation statistics. Dark-field coefficients of the different groups graded with the Fleischner Society grading system (reader median) were compared using the Wilcoxon Mann-Whitney U test. Because of the small number of participants, the confluent emphysema and advanced destructive emphysema groups were combined. We performed receiver operating characteristic curve analysis to evaluate the dark-field coefficient as a classifier once for the presence of at least trace emphysema and once for the presence of at least mild emphysema. We further used the Youden index to determine optimum cutoff values for either task.

Results

Participants

After all exclusions, we studied 83 participants (52 men, 31 women). The average age of the participants in this study was 65 years ± 12 [standard deviation], and the average weight was 74 kg ± 15 (Table 1).

Table 1: Participant Characteristics

Table 1:

Dark-Field Chest Radiography Qualitative Analysis

Figure 3 shows chest radiograph, dark-field chest radiograph, and emphysema projections from CT scans of a 33-year-old man with no pulmonary emphysema (emphysema index = 0.1%) and a 65-year-old man with pulmonary emphysema (emphysema index = 26%). In attenuation-based images, strongly attenuating anatomic structures like the bones, heart, and upper abdominal organs generate a high signal (from a high amount of photon attenuation), while the lung has a limited attenuation (from a low amount of photon attenuation), as it consists mainly of air (Fig 3). However, for dark-field chest radiography, the lung (alveolar microstructure) generates a high signal while bones and solid organs generate little to no signal (Fig 3). In attenuation-based chest radiographs, participants with pulmonary emphysema often have secondary signs of hyperinflation (eg, flattened hemidiaphragms, irregular parenchymal radiolucency, and a paucity of blood vessels) (Fig 3). On the other hand, for dark-field chest radiography, participants with pulmonary emphysema exhibit an overall reduced signal intensity and a rather inhomogeneous patchy appearance compared with the homogeneous high signal intensity seen in healthy participants (Fig 3). The local signal reduction in participants with pulmonary emphysema showed correspondence with the emphysema pattern on CT-based projections (Figs 34). Figure 5 shows results for participants with an emphysema index below 6%. The dark-field chest radiograph shows reduced dark-field signal and an inhomogeneous signal texture.

(A, D) Dark-field radiographs, (B, E) attenuation-based radiographs,                         and (C, F) projections of CT-based emphysema quantification in a healthy                         33-year-old man with an emphysema index of 0.1% (A–C) and a                         65-year-old man with an emphysema index of 26% (D–F). The same window                         settings were applied within the respective modality. In B, no abnormalities                         are apparent. In E, flattened hemidiaphragms and an irregular area of                         radiolucency are visible. While the dark-field chest radiograph of the                         healthy subject with no emphysema in the CT-based emphysema projection                         exhibits a strong homogeneous dark-field signal, the dark-field signal                         intensity of the subject with pulmonary emphysema appears decreased overall                         and exhibits an inhomogeneous patchy pattern, corresponding well to the                         emphysema intensity in the CT-based projection (red overlay, F).

Figure 3: (A, D) Dark-field radiographs, (B, E) attenuation-based radiographs, and (C, F) projections of CT-based emphysema quantification in a healthy 33-year-old man with an emphysema index of 0.1% (A–C) and a 65-year-old man with an emphysema index of 26% (D–F). The same window settings were applied within the respective modality. In B, no abnormalities are apparent. In E, flattened hemidiaphragms and an irregular area of radiolucency are visible. While the dark-field chest radiograph of the healthy subject with no emphysema in the CT-based emphysema projection exhibits a strong homogeneous dark-field signal, the dark-field signal intensity of the subject with pulmonary emphysema appears decreased overall and exhibits an inhomogeneous patchy pattern, corresponding well to the emphysema intensity in the CT-based projection (red overlay, F).

Attenuation-based (top row) and dark-field (middle row) radiographs                         and projections of CT-based emphysema quantification (bottom row) in five                         example participants with a significant emphysema index (EI) of 6% or more.                         The same window and level settings were applied within each respective                         modality. Signs of hyperinflation can be observed in attenuation-based                         images of participants with high emphysema severities. Areas of decreased                         dark-field signal correspond well to emphysematous areas on the CT-based                         emphysema projections. In B, the participant has a right-sided port catheter                         in place.

Figure 4: Attenuation-based (top row) and dark-field (middle row) radiographs and projections of CT-based emphysema quantification (bottom row) in five example participants with a significant emphysema index (EI) of 6% or more. The same window and level settings were applied within each respective modality. Signs of hyperinflation can be observed in attenuation-based images of participants with high emphysema severities. Areas of decreased dark-field signal correspond well to emphysematous areas on the CT-based emphysema projections. In B, the participant has a right-sided port catheter in place.

Attenuation (top row) and dark-field (middle row) radiographs and                         projections of CT-based emphysema quantification (bottom row) in four                         participants with an emphysema index (EI) below 6% but a decreased,                         inhomogeneous dark-field signal intensity. The areas of reduced dark-field                         signal intensity correspond well to the areas of emphysema detected on the                         CT scans, even though these emphysema quantifications are not yet considered                         significant. The same window and level settings were applied within the                         respective modality for all participants.

Figure 5: Attenuation (top row) and dark-field (middle row) radiographs and projections of CT-based emphysema quantification (bottom row) in four participants with an emphysema index (EI) below 6% but a decreased, inhomogeneous dark-field signal intensity. The areas of reduced dark-field signal intensity correspond well to the areas of emphysema detected on the CT scans, even though these emphysema quantifications are not yet considered significant. The same window and level settings were applied within the respective modality for all participants.

Quantitative Analysis

The dark-field coefficient was negatively correlated with the CT-based emphysema index for all participants (r = –0.54, P < .001) and for those participants with an emphysema index greater than or equal to 6% only (r = –0.79, P < .01) (Fig 6). The average dark-field coefficient of the trace emphysema group was similar to that of the absent control group (2.5 m–1 ± 0.4 vs 2.6 m–1 ± 0.4, respectively; P = .34). The average dark-field coefficient was lower in the other Fleischner-grade groups than in the control group (mild group, 2.2 m–1 ± 0.4 [P = .02]; moderate group, 2.1 m–1 ± 0.7 [P = .01]; confluent or advanced destructive group, 1.3 m–1 ± 0.6 [P < .001]) (Table 2). The area under the receiver operating characteristic curve was 0.68 for detection of at least trace emphysema and 0.79 for detection of at least mild emphysema. The optimum cutoff value in dark-field coefficients, below which a participant was classified as having at least trace emphysema was 2.4 m–1 (mild emphysema, 2.3 m–1). The Youden index for trace emphysema detection was J = 0.38 (mild emphysema, J = 0.51). Applying those cutoff values for the detection of emphysema grades of trace or higher, sensitivity, specificity, and accuracy were 0.69, 0.69, and 0.69, respectively. For the detection of emphysema grades of mild or higher, sensitivity, specificity, and accuracy were 0.78, 0.73, and 0.75, respectively.

Dark-field coefficient and CT-based emphysema evaluation of 83                         participants. The dark-field coefficient was calculated by normalizing the                         integral of the dark-field signal of each subject’s lung area with                         their lung volume. (A) Comparison of dark-field coefficient with the                         emphysema index from quantitative CT evaluation. There was moderate                         correlation (r = –0.54, P < .001) between dark-field                         coefficient and emphysema index for all study participants. For participants                         with an emphysema index greater than 6%, Spearman correlation between                         dark-field coefficient and emphysema index was –0.79 (P <                         .01). (B) Comparison of dark-field coefficient and visual emphysema grades                         determined from CT assessment according to the Fleischner scale. The symbols                         indicate the significance of P values from Wilcoxon Mann-Whitney U tests of                         each group with the absent group: ° = P > .05, ★ = P                         < .05. (C) Receiver operating characteristic curve shows capability                         of the dark-field coefficient to detect at least trace emphysema (blue, AUC                         = 0.68) and at least mild emphysema (red, AUC = 0.79).

Figure 6: Dark-field coefficient and CT-based emphysema evaluation of 83 participants. The dark-field coefficient was calculated by normalizing the integral of the dark-field signal of each subject’s lung area with their lung volume. (A) Comparison of dark-field coefficient with the emphysema index from quantitative CT evaluation. There was moderate correlation (r = –0.54, P < .001) between dark-field coefficient and emphysema index for all study participants. For participants with an emphysema index greater than 6%, Spearman correlation between dark-field coefficient and emphysema index was –0.79 (P < .01). (B) Comparison of dark-field coefficient and visual emphysema grades determined from CT assessment according to the Fleischner scale. The symbols indicate the significance of P values from Wilcoxon Mann-Whitney U tests of each group with the absent group: ° = P > .05, * = P < .05. (C) Receiver operating characteristic curve shows capability of the dark-field coefficient to detect at least trace emphysema (blue, AUC = 0.68) and at least mild emphysema (red, AUC = 0.79).

Table 2: Statistical Evaluation of Dark-Field Coefficients

Table 2:

Discussion

In this study, we evaluated the qualitative and quantitative characteristics of x-ray dark-field images in participants with pulmonary emphysema. Compared with images in healthy participants, dark-field chest radiographs in participants with emphysema had lower dark-field signal intensity and an inhomogeneous patchy appearance of the lungs. The locations of focal signal intensity loss on dark-field images corresponded well with emphysematous areas found on CT images. The dark-field coefficient (m–1) was negatively correlated with the quantitative CT-based emphysema index (r = –0.54, P < .001). Participants with Fleischner Society emphysema grades of mild (dark-field coefficient, 2.22 m–1 ± 0.37; P = .02), moderate (2.11 m–1 ± 0.65, P = .01), and confluent or advanced destructive (1.28 m–1 ± 0.62, P < .001) exhibited a lower dark-field coefficient when compared with participants without pulmonary emphysema (2.55 m–1 ± 0.43).

Previous works have already investigated dark-field chest radiography in humans. Gassert et al (23) have described the characteristics of dark-field chest radiography in a cohort of 40 healthy humans and introduced the quantitative dark-field coefficient. They found no significant correlation between the dark-field coefficient and the parameters sex, age, weight, and height. Willer et al (22) investigated dark-field chest radiography in participants with COPD and showed that dark-field chest radiography can be used to detect and grade emphysema in a reader study. However, in that study, dark-field signal was not analyzed quantitatively. Our study applies the quantitative dark-field coefficient introduced by Gassert et al (23) to participants with emphysema.

In this study, emphysematous changes in the lung induced characteristic changes in dark-field images. One effect that was observed was lower signal intensity in participants with emphysema, likely due to fewer alveoli and thus fewer tissue-air interfaces. Additionally, compared with the homogeneous dark-field signal intensity of healthy lungs, the dark-field appearance in participants with emphysema was rather inhomogeneous and patchy due to focal decreases of alveolar density. These findings are consistent with the findings of a small-animal study by Meinel et al (16), who reported a clear visual difference between healthy and emphysematous mice. They also found that the dark-field signal intensity of the lung parenchyma in emphysematous mice was reduced by 25% and was less homogeneous. Similar findings in emphysematous mice were also described by Hellbach et al (17).

Both the CT-based visual emphysema grades and the quantitative emphysema index in CT scans showed results consistent with the quantitative dark-field coefficient, suggesting that quantitative dark-field imaging may be a useful tool in the assessment of pulmonary emphysema. Especially for the detection of mild emphysema, the receiver operating characteristic curve analysis showed that the dark-field coefficient may be a useful quantitative measure, with an area under the receiver operating characteristic curve of 0.79 and an optimum cutoff value of 2.3 m–1. For detection of the earlier-stage trace emphysema, specificity was lower due to a large variation in dark-field coefficients and the large overlap of dark-field coefficient in the absent and trace groups. While dark-field chest radiographs do appear much improved over attenuation radiographs in the detection of moderate diffuse emphysema, more severe emphysema stages are often also visible on attenuation images due to secondary changes, such as flattening of the diaphragm. In these cases, the advantage of dark-field imaging is mainly that focal signal losses may allow for the localization of emphysematous changes. One posteroanterior dark-field image acquisition exposes the patient to 0.037 mSv of radiation (25), which is substantially less (about 2%) than low-dose chest CT. At that low dose, it provides a fairly high sensitivity of 0.78 for detection of mild emphysema, with the CT-based Fleischner grading as the reference standard. While spirometry does not apply any dose and is often used in daily clinical routines, it crucially depends on the operator and cooperation of the patient (27). A previous study has shown that 23% of smokers without spirometric impairment were found to have at least mild emphysema at CT according to the Fleischner scale (2), indicating a low sensitivity of spirometry in detection of mild emphysema. We found that the dark-field coefficient was able to depict mild emphysema; thus, dark-field radiography might close this diagnostic gap and allow the detection of emphysema in early stages at a low dose. In contrast to spirometry, dark-field chest radiography additionally provides image-based information for emphysema localization. Dark-field radiography generates perfectly aligned attenuation-based and dark-field radiographs from one acquisition. Because the attenuation image shows the attenuating anatomic structures and the dark-field image gives information on the alveolar integrity, information from the two complementary modalities might improve the accuracy of emphysema assessment. However, further studies are needed to determine the utility of x-ray dark-field imaging in the detection of emphysema.

X-ray dark-field imaging also allowed for localization of emphysema. Visually, regions of reduced dark-field signal corresponded well to regions of emphysema on CT-based projections. Also, while there was a correlation between the dark-field coefficient and CT-based emphysema assessment, the mean dark-field signal of participants with an emphysema index below 6% covered a wide range of dark-field coefficients. The dependence of the total dark-field signal on the patient’s size is taken into account by the normalization of the dark-field signal with the lung volume, providing the dark-field coefficient. This helps ensure that differences in the dark-field coefficients can be attributed to differences in the alveolar microstructure.

There were limitations to this pilot study. This initial study included a small number of participants with or without emphysema. We did not study other pulmonary diseases, which may also benefit from the use of this method. The dark-field chest radiography methods of emphysema quantification have not been assessed for precision or repeatability and are not likely to be as robust as those currently used for chest CT.

In conclusion, we found that pulmonary emphysema leads to a reduced signal intensity on dark-field chest radiographs, with relatively low radiation exposure. This suggests the potential of the method for diagnosis of chronic obstructive pulmonary disease and other lung diseases that alter the alveolar interfaces in the lung.

Disclosures of conflicts of interest: T.U. disclosed no relevant relationships. F.T.G. disclosed no relevant relationships. M.F. disclosed no relevant relationships. K.W. disclosed no relevant relationships. W.N. disclosed no relevant relationships. P.B. disclosed no relevant relationships. R.C.S. disclosed no relevant relationships. T.K. employee of Philips Innovative Technologies, an affiliate of Royal Philips; inventor or co-inventor of several patents and patent applications related to dark-field imaging, with the assignee of all these being Philips. J.H.B. disclosed no relevant relationships. A.A.F. disclosed no relevant relationships. A.P.S. disclosed no relevant relationships. M.R.M. disclosed no relevant relationships. F.P. disclosed no relevant relationships. D.P. disclosed no relevant relationships.

Acknowledgments

We gratefully acknowledge Andre Braunagel, Hanns-Ingo Maack, Hendrik van der Heijden, Klaus-Jürgen Engel, Bernd Lundt, Sven Prevrhal, Karsten Rindt, Pascal Meyer, Jens von Berg, and Jürgen Mohr for assistance during hardware and software development for the demonstrator system, and Felix Meurer, Yannik Leonhardt, Christina Müller-Leisse, Martin Renz, Nadja Meissner, and Angelika Kammermayer for help with participant handling.

Author Contributions

Author contributions: Guarantors of integrity of entire study, T.U., A.A.F., F.P., D.P.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, T.U., F.T.G., W.N., T.K., A.A.F., M.R.M., F.P., D.P.; clinical studies, T.U., F.T.G., M.F., K.W., J.H.B., A.A.F., D.P.; statistical analysis, T.U., F.T.G., M.F., K.W., J.H.B., D.P.; and manuscript editing, T.U., F.T.G., M.F., K.W., T.K., J.H.B., A.A.F., A.P.S., F.P., D.P.

* T.U. and F.T.G. contributed equally to this work.

** F.P. and D.P. are co-senior authors.

Supported by the European Research Council (AdG 695045); the Federal Ministry of Education and Research; the Free State of Bavaria under the Excellence Strategy of the Federal Government and the Länder; the German Research Foundation (GRK2274); the Institute for Advanced Study, Technische Universität München; and the Karlsruhe Nano Micro Facility, a Helmholtz Research Infrastructure at Karlsruhe Institute of Technology.

References

  • 1. O’Brien C, Guest PJ, Hill SL, Stockley RA. Physiological and radiological characterisation of patients diagnosed with chronic obstructive pulmonary disease in primary care. Thorax 2000;55(8):635–642. Crossref, MedlineGoogle Scholar
  • 2. Lynch DA, Moore CM, Wilson C, et al. CT-based Visual Classification of Emphysema: Association with Mortality in the COPDGene Study. Radiology 2018;288(3):859–866. LinkGoogle Scholar
  • 3. Johns DP, Walters JA, Walters EH. Diagnosis and early detection of COPD using spirometry. J Thorac Dis 2014;6(11):1557–1569. MedlineGoogle Scholar
  • 4. Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Pulmonary Disease. Global Initiative for Chronic Obstructive Lung Disease. https://goldcopd.org/wp-content/uploads/2019/12/GOLD-2020-FINAL-ver1.2-03Dec19_WMV.pdf. Updated 2020. Google Scholar
  • 5. Müller NL, Staples CA, Miller RR, Abboud RT. “Density mask”. An objective method to quantitate emphysema using computed tomography. Chest 1988;94(4):782–787. Crossref, MedlineGoogle Scholar
  • 6. Wang Z, Gu S, Leader JK, et al. Optimal threshold in CT quantification of emphysema. Eur Radiol 2013;23(4):975–984. Crossref, MedlineGoogle Scholar
  • 7. den Harder AM, de Boer E, Lagerweij SJ, et al. Emphysema quantification using chest CT: influence of radiation dose reduction and reconstruction technique. Eur Radiol Exp 2018;2(1):30. Crossref, MedlineGoogle Scholar
  • 8. Lynch DA, Austin JH, Hogg JC, et al. CT-Definable Subtypes of Chronic Obstructive Pulmonary Disease: A Statement of the Fleischner Society. Radiology 2015;277(1):192–205. LinkGoogle Scholar
  • 9. National Lung Screening Trial Research Team; Aberle DR, Adams AM, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med 2011;365(5):395–409. Crossref, MedlineGoogle Scholar
  • 10. Kroft LJM, van der Velden L, Girón IH, Roelofs JJH, de Roos A, Geleijns J. Added Value of Ultra-low-dose Computed Tomography, Dose Equivalent to Chest X-Ray Radiography, for Diagnosing Chest Pathology. J Thorac Imaging 2019;34(3):179–186. Crossref, MedlineGoogle Scholar
  • 11. Pfeiffer F, Bech M, Bunk O, et al. Hard-X-ray dark-field imaging using a grating interferometer. Nat Mater 2008;7(2):134–137. Crossref, MedlineGoogle Scholar
  • 12. Schleede S, Meinel FG, Bech M, et al. Emphysema diagnosis using X-ray dark-field imaging at a laser-driven compact synchrotron light source. Proc Natl Acad Sci U S A 2012;109(44):17880–17885. Crossref, MedlineGoogle Scholar
  • 13. Yaroshenko A, Hellbach K, Yildirim AO, et al. Improved In vivo Assessment of Pulmonary Fibrosis in Mice using X-Ray Dark-Field Radiography. Sci Rep 2015;5(1):17492. Crossref, MedlineGoogle Scholar
  • 14. Yaroshenko A, Pritzke T, Koschlig M, et al. Visualization of neonatal lung injury associated with mechanical ventilation using x-ray dark-field radiography. Sci Rep 2016;6(1):24269. Crossref, MedlineGoogle Scholar
  • 15. Scherer K, Yaroshenko A, Bölükbas DA, et al. X-ray Dark-field Radiography - In-Vivo Diagnosis of Lung Cancer in Mice. Sci Rep 2017;7(1):402. Crossref, MedlineGoogle Scholar
  • 16. Meinel FG, Yaroshenko A, Hellbach K, et al. Improved diagnosis of pulmonary emphysema using in vivo dark-field radiography. Invest Radiol 2014;49(10):653–658. Crossref, MedlineGoogle Scholar
  • 17. Hellbach K, Yaroshenko A, Meinel FG, et al. In Vivo Dark-Field Radiography for Early Diagnosis and Staging of Pulmonary Emphysema. Invest Radiol 2015;50(7):430–435. Crossref, MedlineGoogle Scholar
  • 18. Gromann LB, De Marco F, Willer K, et al. In-vivo X-ray Dark-Field Chest Radiography of a Pig. Sci Rep 2017;7(1):4807. Crossref, MedlineGoogle Scholar
  • 19. De Marco F, Willer K, Gromann LB, et al. Contrast-to-noise ratios and thickness-normalized, ventilation-dependent signal levels in dark-field and conventional in vivo thorax radiographs of two pigs. PLoS One 2019;14(6):e0217858. Crossref, MedlineGoogle Scholar
  • 20. Willer K, Fingerle AA, Gromann LB, et al. X-ray dark-field imaging of the human lung-A feasibility study on a deceased body. PLoS One 2018;13(9):e0204565. Crossref, MedlineGoogle Scholar
  • 21. Sauter AP, Andrejewski J, De Marco F, et al. Optimization of tube voltage in X-ray dark-field chest radiography. Sci Rep 2019;9(1):8699. Crossref, MedlineGoogle Scholar
  • 22. Willer K, Fingerle AA, Noichl W, et al. X-ray dark-field chest imaging for detection and quantification of emphysema in patients with chronic obstructive pulmonary disease: a diagnostic accuracy study. Lancet Digit Health 2021;3(11):e733–e744. Crossref, MedlineGoogle Scholar
  • 23. Gassert FT, Urban T, Frank M, et al. X-ray Dark-Field Chest Imaging: Qualitative and Quantitative Results in Healthy Humans. Radiology 2021;301(2):389–395. LinkGoogle Scholar
  • 24. Pfeiffer FWT, Bunk O, David C. Phase retrieval and differential phase-contrast imaging with low-brilliance X-ray sources. Nat Phys 2006;2(4):258–261. CrossrefGoogle Scholar
  • 25. Frank M, Urban T, Willer K, et al. Dosimetry on first clinical dark-field chest radiography. Med Phys 2021;48(10):6152–6159. Crossref, MedlineGoogle Scholar
  • 26. Pierce RJ, Brown DJ, Holmes M, Cumming G, Denison DM. Estimation of lung volumes from chest radiographs using shape information. Thorax 1979;34(6):726–734. Crossref, MedlineGoogle Scholar
  • 27. Walters JA, Hansen EC, Johns DP, Blizzard EL, Walters EH, Wood-Baker R. A mixed methods study to compare models of spirometry delivery in primary care for patients at risk of COPD. Thorax 2008;63(5):408–414. Crossref, MedlineGoogle Scholar

Article History

Received: Aug 12 2021
Revision requested: Sept 13 2021
Revision received: Oct 18 2021
Accepted: Oct 28 2021
Published online: Jan 11 2022
Published in print: Apr 2022