AJR ARRS Member Benefits
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kim, H. H.
Right arrow Articles by Koomen, M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kim, H. H.
Right arrow Articles by Koomen, M. A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?
DOI:10.2214/AJR.05.0187
AJR 2006; 187:47-50
© American Roentgen Ray Society

Original Research

Comparison of Calcification Specificity in Digital Mammography Using Soft-Copy Display Versus Screen-Film Mammography

Hak Hee Kim1, Etta D. Pisano2, Elodia B. Cole2, Michael R. Jiroutek3, Keith E. Muller4, Yuanshui Zheng2, Cherie M. Kuzmiak5 and Marcia A. Koomen5

1 Department of Radiology, University of Ulsan, Asan Medical Center, 388-1, Seoul 138-736, South Korea.
2 Department of Radiology and Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, 27599.
3 Salix Pharmaceuticals, Inc., Morrisville, North Carolina.
4 Department of Biostatistics, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina.
5 Department of Radiology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina.

Received February 3, 2005; accepted after revision May 3, 2005.

 
Address correspondence to E. D. Pisano.


Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
OBJECTIVE. The purpose of this study was to compare specificity in the interpretation of calcifications in soft-copy reviewing of digital mammograms versus hard-copy reviewing of screen-film mammograms.

MATERIALS AND METHODS. A total of 130 consecutive cases with calcifications (44 malignant and 86 benign) that had been evaluated with needle or surgical biopsy were collected. Both screen-film mammography and soft-copy digital mammography were obtained in the same patients under existing research protocols using Fischer Imaging's SenoScan (n = 71), Lorad's digital mammography system (n = 35), and GE Healthcare's Senographe 2000D (n = 24). Eight trained radiologists scored all lesions—cropped or masked to display just the region of interest—both on screen-film and soft-copy digital mammography with a month between reviews to reduce the effects of learning and memory. A 5-point malignancy scale was used, with 1 as definitely not, 2 as probably not, 3 as possibly, 4 as probably, and 5 as definitely. Reviewers were randomly assigned condition order, and images within each condition were randomly ordered. Repeated measures analysis of variance was used to test for differences between conditions in specificity computed via nonparametric receiver operating characteristic (ROC) study separately for each reviewer and condition.

RESULTS. Across all reviewers, the mean specificity for 1 or 2 versus 3, 4, or 5 was 0.803 for screen-film mammography (range, 0.413-0.938; SD ± 0.166) and 0.833 for soft-copy image (range, 0.375-0.951; SD ± 0.187). Although not statistically significant (Student's t test p values from 0.19 to 0.99 across all cut points), numeric values of specificity were consistently higher for soft-copy versus screen-film mammography. No statistical significance in specificity was seen using all possible cut points in the 5-point scale, although the primary analysis used the cutpoint for differentiation between benign and malignant cases as 1 or 2 versus 3, 4, or 5.

CONCLUSION. No statistically significant difference was shown in specificity achievable using soft-copy digital versus screen-film mammography in this study.

Keywords: breast • comparative studies • diagnostic radiology • digital images • mammography • observer performance


Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Screen-film mammography has long been the standard technique for breast imaging. However, it has inherent limitations, such as low detective quantum efficiency and the high difficulty of postprocessing after obtaining an image. Digital mammography has several advantages over screen-film mammography because digital mammography systems directly quantify X-ray photons and decouple the process of X-ray photon detection from image display. Digital images can be processed by a computer and displayed in multiple formats, either on printed film or through soft-copy display systems. However, once digital mammograms are printed and displayed on laser films, the flexibility of this system is lost. Soft-copy digital mammography is flexible, allowing online contrast manipulation with real-time adjustment. Its ability to manipulate the contrast of an image after acquisition could lead to decreases in requests by radiologists for repeat images or callbacks for additional imaging solely on the basis of low contrast between healthy and abnormal tissue (i.e., women with dense breast tissue).

In this study, we investigated using soft-copy display for characterization of lesions from screening mammograms of women with dense breasts. Specifically, the objective of this study was to compare the specificity of screen-film and soft-copy image mammography for microcalcifications and to determine whether a soft-copy image provides improved specificity compared with screen-film mammography.


Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
A total of 130 consecutive cases with histologically proven calcifications by needle or surgical biopsy were included in this study. The only inclusion criterion was that the lesions depicted were calcifications that had undergone biopsy. Both screen-film and soft-copy digital mammography images were acquired in the same patients. Two standard views of each breast were obtained with each technique. Cases were accrued under clinical trials using three different full-field digital mammographic devices: the Fischer Imaging SenoScan was used for 71 of the patients, the Lorad digital mammography system was used for 35 of the patients, and the GE Healthcare Senographe 2000D was used for 24 of the patients. Images obtained with the Fischer system were 3,072 x 4,800 pixels with 50-µm pixels. The Lorad system images were 4,800 x 6,400 pixels with 40-µm pixels and the GE Healthcare system images were 1,800 x 2,304 pixels with 100-µm pixels. The Lorad and GE Healthcare systems produced images with 14 bits per pixel, and the Fischer system had 12 bits per pixel.

Soft-copy images were cropped so that just the region of interest was shown, and these subimages were displayed, in both craniocaudal and mediolateral oblique views, using high-luminance, high-spatial-resolution monitors of 2,048 x 2,506 pixels (model 1654, Orwin Associates). Screen-film mammograms were masked to show only the same calcifications and were displayed on a light box. An experienced radiologist, who did not participate in the scoring of the images, annotated and masked the films.

Eight trained radiologists, all qualified mammographic interpreters under the U.S. Federal Mammography Quality Standards Act (MQSA), independently participated as interpreters in this study. The interpreters all review mammograms as part of their everyday clinical practice of breast imaging. The performance task was to provide a probability of malignancy for each group of calcifications based on the perceived characteristics of the calcifications from screening mammograms. Calcifications were visible within the images presented for each case. The interpreters were shown only regions of the mammograms containing calcifications so that they could focus solely on the characterization task. Each interpreter scored all lesions, using both soft-copy digital and screen-film mammography. At least a month passed between interpretations in each condition, to reduce the effects of learning and memory. Interpreters were randomly assigned condition order, and images within each condition were randomly ordered within a technique. A 5-point malignancy scale was used, with 1 defined as definitely not malignant, 2 as probably not malignant, 3 as possibly malignant, 4 as probably malignant, and 5 as definitely malignant. The BI-RADS standard scale for likelihood of cancer classification was not used because it is not a continuous scale and this study did not focus on the screening task per se (callback vs no callback). A rating of 1 or 2 on our malignancy scale was called benign; and 3, 4, or 5 was called malignant. The scale enabled the radiologists to give only their estimates of the likelihood of malignancy, allowing them to assume all cases reviewed would undergo biopsy.

The radiologists were able to adjust window and level and to magnify (x2 power) each image interactively for soft-copy image. Appropriate masking of the film and viewbox was used for screen-film mammography. Radiologists were provided with a magnifying glass (x2 power) for screen-film mammography interpretation. No prior films, patient history, or diagnostic views (magnification images or spot radiographs) were provided to the interpreters. A research associate presented the cases to each interpreter in random order.

Repeated measures analysis of variance was used separately for each interpreter and technique to test for differences in specificity between conditions computed via nonparametric receiver operating characteristic (ROC) study. To address our objectives while weighting the type 1 error rate to reflect the tests of most interest, a traditional stepdown approach to the repeated measures analysis of variance was conducted with the alpha level divided as follows. First, we tested for the significance of the cut point by technique interaction, which asks if the difference between soft-copy and screen-film mammographies is different across the cut points, at the 0.02 {alpha}-level. Given a nonsignificant interaction, we then performed stepdown tests to the main effect test of technique, which tests for the difference between the soft-copy and the screen-film mammography averaged across the cut point, at the 0.02 {alpha}-level. Because the main effect of the cut point was not of interest, this test was ignored. Of secondary interest, regardless of the significance of the interaction, were the four separate Student's t tests comparing the soft-copy and screen-film mammographies at each cut point. These Student's t tests were essentially the stepdown tests for a significant technique by cut point interaction, but they were conducted whether the interaction was significant or not. Each Student's t test was conducted at the 0.01/4 equal to 0.0025 level. Although we were interested in the soft-copy image only if it was superior to the screen-film mammography image, the possibility existed that it would be worse than the screen-film mammography image. Therefore, all of the tests were two-sided.


Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Of our 130 cases, 44 contained biopsy-proven malignant calcifications and 86 contained biopsy-proven benign calcifications.

Table 1 shows the mean specificity for each cut point, which is averaged across all interpreters. For all interpreters, the mean specificity for malignancy scores of 1 or 2 versus 3, 4, or 5 was 0.803 for screen-film mammography (range, 0.413-0.938; SD ± 0.166) and 0.833 for soft-copy digital mammography (range, 0.375-0.951; SD ± 0.187). As a visual aid, Figure 1 plots the mean specificities from Table 1 versus cut point by technique. Note that the cut point value 1 refers to the 1 versus ≥ 2 split, the cut point value 2 refers to the ≤ 2 versus 3 3 split, the cut point value 3 refers to the ≤ 3 versus ≥ 4 split, and the cut point value 4 refers to the ≤ 4 versus 5 split. The separation of the lines at the second and third cut points shows that the soft-copy image has a nonsignificantly higher specificity at these points, which are the two cut points of the most clinical importance.


View this table:
[in this window]
[in a new window]

  		TABLE 1: Mean Specificities (SD) for Each Cut Point, Averaged Across Interpreters

 

Figure 1
View larger version (5K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 1 —Mean specificity versus cut point by technique. Solid line = screen-film mammography, dotted line = soft-copy digital mammography.

 

The test of the cut point by technique interaction resulted in an F-test statistic of 0.74 with a corresponding p value of 0.57, suggesting that the difference between soft-copy and screen-film mammographies is not statistically different across the cut points. The main effect test of technique resulted in an F-test statistic of 1.93 with a corresponding p value of 0.21, suggesting that the difference between soft-copy and screen-film mammographies averaged across the cut point is not statistically significant. Last, Table 2 shows the results of the four separate Student's t tests comparing screen-film and soft-copy mammographies at each cut point. The lack of significance of these four tests suggests that no statistically significant difference exists between techniques for any of the individual cut points considered, although the primary analysis used the cut point for differentiation between benign and malignant as 1 or 2 versus 3, 4, or 5. Although not statistically significant (Student's t test p values from 0.19 to 0.99 across all cut points), the numeric values of specificity were consistently higher for soft-copy digital mammography than for screen-film mammography (Table 1).


View this table:
[in this window]
[in a new window]

  		TABLE 2: Results of the Four Separate Student's t Tests Comparing Screen-Film Mammography and Soft-Copy Digital Mammography at Each Cut Point

 

Interpreter variability did not significantly affect the results of this study.


Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Characterization of breast calcifications is one of the main clinical tasks of mammography and typically requires compression and spot magnification views before a final BIRADS diagnosis is reached. Although it is possible to improve visibility of calcifications through manipulation of contrast resolution in some instances, contrast resolution is not the sole factor influencing lesion visibility. Spatial resolution is also important in diagnosing calcifications and other small features in the image, which is why compression and spot magnification views with higher spatial resolution contained within smaller areas of the image are the imaging tools of choice for calcifications characterization. In comparing digital mammography to screen-film mammography, the former has an advantage over the latter because higher contrast resolution is available with the ability to adjust the contrast of the mammograms through use of image processing. However, screen-film mammography has higher spatial resolution than digital mammography, and more detailed image features may be obvious. Several studies show that despite the limiting lower spatial resolution, visibility of calcifications on digital mammography is not significantly different from that on screen-film mammography [1-7]. In addition, pixel sizes in the range studied did not have a strong effect on radiologists' accuracy in the characterization of calcifications [4]. Lewin et al. [6] reported that although screen-film mammography did recall a larger number of cases containing calcifications, the number of cancers manifesting as microcalcifications was the same with both techniques. A higher percentage of digital mammography-only calcification findings were positive at biopsy suggesting that the soft-copy capabilities of digital mammography might allow image manipulations that provide improved visibility of lesion features and give the radiologist more information as a cause for improved specificity [6]. Obenauer et al. [7] reported that the flat-panel digital mammography system was superior to screen-film mammography in image quality, detail visibility, image exposure, and artifacts. Lesion detectability and characterization, according to BI-RADS classifications in digital images, was equal or even superior to that seen with screen-film mammography [7]. They concluded that the better contrast detectability and the ability to do image processing highly compensated for the limitations in spatial resolution.

However, most previous reports have focused on comparing observer performance with screen-film mammography and hardcopy digital mammography. As PACS becomes more universally available, it is expected that mammography will also rapidly convert to soft-copy display. Observer performance in soft-copy display must also be compared with that in screen-film mammography and in hard-copy digital mammography to determine whether digital mammography can completely replace film-based mammography.

The diagnostic accuracy of soft-copy and hard-copy interpretation is likely to be comparable if a high-resolution laser printer and a high-quality workstation with high-spatialand high-contrast-resolution monitors are used [8]. However, when digital mammograms are printed and displayed on laser film, the flexibility of digital imaging is lost because the display parameters must be chosen before printing. All of the available information cannot be optimally displayed in a single presentation. However, soft-copy display is flexible, allowing online contrast manipulation, roaming, and zooming to full resolution.

Our hypothesis was that soft-copy digital mammography provides improved specificity compared with screen-film mammography for the diagnosis of breast calcifications. Although our results were not statistically significant, the numeric values of specificity were consistently higher for soft-copy display than for screen-film mammography, revealing a trend in favor of the soft-copy technique. Recent studies support our results [1, 10]. Kuzmiak et al. [9] reported no statistically significant difference in diagnostic accuracy, including microcalcifications, between magnified screen-film mammography and unmagnified soft-copy digital mammography in breast tissue biopsy specimens.

Many challenges remain as screen-film mammography is converted to soft-copy digital mammography. The interpretation of the soft-copy image may be more time-consuming than that of the hard-copy digital mammography because of the additional digital operations associated with viewing soft-copy images. A recent study also revealed that speed and accuracy of the interpretations of digital mammography using printed-film versus soft-copy display are not significantly different, but soft-copy interpretation is slightly faster [11]. Sensitivity was slightly greater for interpretations with printed film, and specificity was slightly greater for interpretations with soft copy. Interpretation with soft-copy display is likely to be useful with digital mammography and is unlikely to significantly change accuracy or speed. Other challenges are the high costs associated with the required digital infrastructure, data storage and transmission, and developing support for this change from referring clinicians [12].

Despite these disadvantages, the long-term prospects of a filmless environment and soft-copy review seem to be inextricably linked to the future success of digital mammography. For soft-copy image interpretation to become an accepted replacement for screen-film mammography interpretation, its accuracy must be clearly established. Recent studies indicate that no compromise was noted in the diagnostic accuracy of soft-copy digital mammography compared with that of screen-film mammography. Future enhancements—such as computer-aided detection and diagnosis and other new applications such as telemammography, tomosynthesis, contrast subtraction, and dual energy subtraction—could potentially result in further improvements in diagnostic accuracy.

This study design has a few limitations. The first is the relatively small number of cases it analyzed, thereby making it difficult to establish statistical significance for the various digital systems. The insufficient number of cases from each machine type did not allow comparison of the soft-copy performance of various machine types. In addition, the soft-copy system used was not commercially available, but rather a system that allowed display of images from any digital manufacturer. Another limitation is the exclusion of sensitivity from this study, which would give a more complete performance comparison. However, several articles have previously showed that detection sensitivity of calcifications on digital mammography does not differ significantly from that of screen-film mammography [6, 7, 13-15]. The third limitation is that this study did not include available compression and magnification views. However, because this limitation affected both techniques equally, it should not have altered our results. Finally, we did not study the effects of breast parenchymal pattern, patient age, location of the lesions, or the exact interpretation time on our results. The role of these factors will need to be evaluated in future studies.

In conclusion, no statistically significant difference was shown in specificity achievable using soft-copy versus screen-film mammography in this study.


Acknowledgments
 
We thank the International Digital Mammography Development Group for their provision of the cases used in this study.


References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Hermann KP, Obenauer S, Funke M, Grabbe EH. Magnification mammography: a comparison of full-field digital mammography and screen-film mammography for the detection of simulated small masses and microcalcifications. Eur Radiol 2002;12 : 2188-2191[Medline]
  2. Diekmann F, Diekmann S, Bick U, et al. Comparing the visualization of microcalcifications with direct magnification in digital full-field mammography vs. film-screen mammography. Rofo2002; 174:297 -300[Medline]
  3. Yip WM, Pang SY, Yim WS, Kwok CS, ROC curve analysis of lesion detectability on phantoms: comparison of digital spot mammography with conventional spot mammography. Br J Radiol2001; 74:621 -628[Abstract/Free Full Text]
  4. Chan HP, Helvie MA, Petrick N, et al. Digital mammography: observer performance study of the effects of pixel size on the characterization of malignant and benign microcalcifications. Acad Radiol2001; 8:454 -466[CrossRef][Medline]
  5. Fischer U, Baum F, Obenauer S, et al. Comparative study in patients with microcalcifications: full-field digital mammography vs screen-film mammography. Eur Radiol 2002;12 : 2679-2683[Medline]
  6. Lewin JM, Hendrick RE, D'Orsi CJ, et al. Comparison of full-field digital mammography to screen-film mammography for cancer detection: results of 4945 paired examinations. Radiology2001; 218:873 -880[Abstract/Free Full Text]
  7. Obenauer S, Luftner-Nagel S, von Heyden D, Munzel U, Baum F, Grabbe E. Screen film vs full-field digital mammography: image quality, detectability and characterization of lesions. Eur Radiol2002; 12:1697 -1702[CrossRef][Medline]
  8. Funke M, Obenauer S, Hermann KP, Fischer U, Grabbe E. Soft copy versus hard copy findings in digital mammography [in German]. Radiologe 2001;42 : 265-269[CrossRef]
  9. Kuzmiak CM, Millnamow GA, Qaqish B, Pisano ED, Cole EB, Brown ME. Comparison of full-field digital mammography to screen-film mammography with respect to diagnostic accuracy of lesion characterization in breast tissue biopsy specimens. Acad Radiol 2002;12 : 1378-1382
  10. Pisano ED, Gatsonis C, Hendrick E, et al. Diagnostic performance of digital versus film mammography for breast-cancer screening. N Engl J Med 2005; 353:1773 -1783[Abstract/Free Full Text]
  11. Pisano ED, Cole EB, Kistner EO, et al. Interpretation of digital mammograms: comparison of speed and accuracy of soft-copy versus printed-film display. Radiology 2002;223 : 483-488[Abstract/Free Full Text]
  12. Johnson CD. Hard-versus soft-copy interpretation. Radiology 2003;227 : 629-630[Free Full Text]
  13. Skaane P, Skjennald A. Screen-film mammography versus full-field digital mammography with soft-copy reading: randomized trial in a population-based screening program—the Oslo II Study. Radiology 2004;232 : 197-204[Abstract/Free Full Text]
  14. Obenauer S, Hermann KP, Marten K, et al. Soft copy versus hard copy reading in digital mammography. J Digit Imaging2003; 16:341 -344[Medline]
  15. Di Nubila B, Cassano E, Origgi D, et al. Analogic versus digital mammographic examination a radiological study of mammary microcalcifications on 52 surgical samples. Radiol Med (Torino)2003; 106:297 -304

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kim, H. H.
Right arrow Articles by Koomen, M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kim, H. H.
Right arrow Articles by Koomen, M. A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS