Rationale and Objectives
The purpose of this study was to assess the performance of a MicroDose photon-counting full-field digital mammography (PCM) system in comparison to full-field digital mammography (FFDM) for area under the receiver-operating characteristic (ROC) curve (AUC), sensitivity, specificity, and feature analysis of standard-view mammography for women presenting for screening mammography, diagnostic mammography, or breast biopsy.
Materials and Methods
A total of 133 women were enrolled in this study at two European medical centers, with 67 women who had a pre-existing 10–36 months FFDM enrolled prospectively into the study and 66 women who underwent breast biopsy and had screening PCM and diagnostic FFDM, including standard craniocaudal and mediolateral oblique views of the breast with the lesion, enrolled retrospectively. The case mix consisted of 49 cancers, 17 biopsy-benign cases, and 67 normal cases. Sixteen radiologists participated in the reader study and interpreted all 133 cases in both conditions, separated by washout period of ≥4 weeks. ROC curve and free-response ROC curve analyses were performed for noninferiority of PCM compared to FFDM using a noninferiority margin Δ value of 0.10. Feature analysis of the 66 cases with lesions was conducted with all 16 readers at the conclusion of the blinded reads. Mean glandular dose was recorded for all cases.
Results
The AUC for PCM was 0.947 (95% confidence interval [CI], 0.920–0.974) and for FFDM was 0.931 (95% CI, 0.898–0.964). Sensitivity per case for PCM was 0.936 (95% CI, 0.897–0.976) and for FFDM was 0.908 (95% CI, 0.856–0.960). Specificity per case for PCM was 0.764 (95% CI, 0.688–0.841) and for FFDM was 0.749 (95% CI, 0.668–0.830). Free-response ROC curve figures of merit were 0.920 (95% CI, 0.881–0.959) and 0.903 (95% CI, 0.858–0.948) for PCM and FFDM, respectively. Sensitivity per lesion was 0.903 (95% CI, 0.846–0.960) and 0.883 (95% CI, 0.823–0.944) for PCM and FFDM, respectively. The average false-positive marks per image of noncancer cases were 0.265 (95% CI, 0.171–0.359) and 0.281 (95% CI, 0.188–0.374) for PCM and FFDM, respectively. Noninferiority P values for AUC, sensitivity (per case and per lesion), specificity, and average false-positive marks per image were all statistically significant ( P < .001). The noninferiority P value for free-response ROC was <.025, from the 95% CI for the difference. Feature analysis resulted in PCM being preferred to FFDM by the readers for ≥70% of the cases. The average mean glandular dose for PCM was 0.74 mGy (95% CI, 0.722–0.759 mGy) and for FFDM was 1.23 mGy (95% CI, 1.199–1.262 mGy).
Conclusions
In this study, radiologist performance with PCM was not inferior to that with conventional FFDM at an average 40% lower mean glandular dose.
Several published studies have demonstrated that digital mammography systems have significantly lower doses than screen-film mammography, with the slot-scanning systems having the lowest doses at comparable exposures across all breast densities. Although numerous technologies have been used for breast cancer screening, the gold standard since the late 1970s has been x-ray-based mammography . Technological advancements have led to monumental improvements in the detection and diagnostic accuracy of mammography in the past 30 years, culminating in the introduction and dissemination of full-field digital mammography (FFDM) in the past 10 years.
The transition to digital mammography was brought about by the desire to improve the sensitivity of x-ray mammography and decrease the amount of unnecessary radiation exposure that was experienced in film mammographic acquisition, especially in women with dense breasts. In the Digital Mammographic Imaging Screening Trial (DMIST), the diagnostic accuracy of digital and film systems was shown to be comparable, with digital mammography showing improved performance over film mammography for women with dense breasts, those aged <50 years, and those who were premenopausal or perimenopausal . Digital mammography systems tested in DMIST on average had a 22% lower mean glandular dose (MGD) than screen-film mammography systems used in DMIST . The slot-scanning FFDM system had the lowest MGD in comparison to screen-film mammography. Slot-scanning systems have been shown to have lower MGDs than conventional screen-film mammography systems . The photon-counting FFDM (PCM) system combines photon counting for high detective quantum efficiency with scanning multislit technology for efficient scatter rejection. A potential for 40% to 60% dose reduction with maintained physical image quality has been reported for the system . The aim of this study was to evaluate the clinical effectiveness of a PCM system in comparison to conventional FFDM.
Materials and methods
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Multireader, Multicase (MRMC) Task
Get Radiology Tree app to read full this article<
Feature Analysis Task
Get Radiology Tree app to read full this article<
Statistical Methods
MRMC
Get Radiology Tree app to read full this article<
Feature analysis
Get Radiology Tree app to read full this article<
Results
Case-level Analysis
AUC
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Sensitivity
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Specificity
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Lesion-level Analysis
Get Radiology Tree app to read full this article<
Table 1
FROC, Sensitivity, and False-positive Rate per Lesion Analyses Averaged across All Readers
PCM FFDM Difference Noninferiority P Value Secondary aims FROC curve figure of merit, POM 0.920 (0.881 to 0.959) 0.903 (0.858 to 0.948) 0.017 (−0.007 to 0.041) NA Sensitivity, per lesion, BI-RADS 0.903 (0.846 to 0.960) 0.883 (0.823 to 0.944) 0.019 (−0.019 to 0.058) <.001 Average false-positive marks per image, only cases without cancer, BI-RADS 0.265 (0.171 to 0.359) 0.281 (0.188 to 0.374) −0.016 (−0.061 to 0.029) <.001
BI-RADS, Breast Imaging Reporting and Data System; FFDM, full-field digital mammography; FROC, free-response receiver-operating characteristic; NA, not applicable; PCM, photon-counting digital mammography; POM, probability of malignancy.
Ninety-five percent confidence intervals are presented in parentheses.
Get Radiology Tree app to read full this article<
Feature Analysis
Get Radiology Tree app to read full this article<
Table 2
Proportion of Noninferior Conspicuity Ratings
View/Reader ID PCM Noninferior ∗ Minimum Conspicuity † Maximum Conspicuity ‡ Skin Line Chest Wall Craniocaudal plus mediolateral oblique 1 101/130 (78%) 103/130 (79%) 132/132 (100%) 132/132 (100%) 2 86/122 (70%) 87/122 (71%) 125/132 (95%) 122/130 (94%) 3 75/124 (60%) 79/124 (64%) 120/132 (91%) 108/132 (82%) 4 73/126 (58%) 76/126 (60%) 131/132 (99%) 19/26 (73%) 5 105/122 (86%) 105/122 (86%) 129/132 (98%) 129/132 (98%) 6 82/124 (66%) 86/124 (69%) 131/132 (99%) 129/132 (98%) 7 94/126 (75%) 96/126 (76%) 40/132 (30%) 36/132 (27%) 8 81/130 (62%) 83/130 (64%) 102/130 (78%) 57/130 (44%) 9 93/122 (76%) 96/122 (79%) 57/132 (43%) 33/132 (25%) 10 100/126 (79%) 102/126 (81%) 130/132 (98%) 127/132 (96%) 11 69/128 (54%) 72/128 (56%) 132/132 (100%) 125/132 (95%) 12 105/130 (81%) 107/130 (82%) 130/132 (98%) 128/130 (98%) 13 84/130 (65%) 86/130 (66%) 130/132 (98%) 118/132 (89%) 14 80/130 (62%) 83/130 (64%) 130/132 (98%) 126/132 (95%) 15 98/126 (78%) 101/126 (80%) 130/132 (98%) 82/126 (65%) 16 89/128 (70%) 91/128 (71%) 129/132 (98%) 129/130 (99%)
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Table 3
Proportion of PCM Cases Scored as Noninferior in Feature Visibility in Comparison to FFDM
Variable CC MLO Difference (CC − MLO) Average Lesion conspicuity (minimum rating) 0.680 (0.597 to 0.762) 0.715 (0.639 to 0.791) −0.042 (−0.113 to 0.029) 0.700 (0.629 to 0.770) Lesion conspicuity (maximum rating) 0.698 (0.618 to 0.779) 0.733 (0.659 to 0.808) −0.042 (−0.112 to 0.029) 0.718 (0.650 to 0.787) Skin line tissue visibility 0.887 (0.776 to 0.999) 0.892 (0.776 to 1.000) −0.004 (−0.015 to 0.007) 0.890 (0.777 to 1.000) Chest wall tissue visibility 0.811 (0.678 to 0.943) 0.798 (0.649 to 0.948) 0.021 (−0.019 to 0.062) 0.799 (0.660 to 0.938)
CC, craniocaudal; FFDM, full-field digital mammography; MLO, mediolateral oblique; PCM, photon-counting digital mammography.
Ninety-five percent confidence intervals are presented in parentheses.
Get Radiology Tree app to read full this article<
Dose
Get Radiology Tree app to read full this article<
Table 4
Average MGD and Breast Thickness by BI-RADS Density Category and Population Cohort
BI-RADS Density PCM FFDM Average MGD per Exposure (mGy) Average Thickness (mm) Average MGD per Exposure (mGy) Average Thickness (mm) Mostly fatty 0.82 75 1.43 72 Scattered fibroglandular 0.72 56 1.24 56 Heterogeneously dense 0.75 50 1.20 49 Extremely dense 0.80 45 1.26 46 Population cohort Normal 0.70 48 1.20 48 Diagnostic 0.84 62 1.31 62 All subjects 0.74 53 1.23 52
BI-RADS, Breast Imaging Reporting and Data System; FFDM, full-field digital mammography; MGD, mean glandular dose; PCM, photon-counting digital mammography.
Get Radiology Tree app to read full this article<
Discussion
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Conclusions
Get Radiology Tree app to read full this article<
Acknowledgments
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
References
1. Hendrick R.E., Pisano E.D., Averbukh A., et. al.: Comparison of acquisition parameters and breast dose in digital mammography and screen-film mammography in the American College of Radiology Imaging Network Digital Mammographic Imaging Screening Trial. AJR Am J Roentgenol 2010; 194: pp. 362-369.
2. Hendrick R.E., Cole E.B., Pisano E.D., et. al.: Accuracy of soft-copy digital mammography versus that of screen-film mammography according to digital manufacturer: ACRIN DMIST retrospective multireader study. Radiology 2008; 247: pp. 38-48.
3. Feig S.A., Shaber G.S., Schwartz G.F., et. al.: Thermography, mammography, and clinical examination in breast cancer screening. Review of 16,000 studies. Radiology 1977; 122: pp. 123-127.
4. Pisano E.D., Gatsonis C., Hendrick E., et. al.: Diagnostic performance of digital versus film mammography for breast-cancer screening. N Engl J Med 2005; 353: pp. 1773-1783.
5. Pisano E.D., Hendrick R.E., Yaffe M.J., et. al.: Diagnostic accuracy of digital versus film mammography: exploratory analysis of selected population subgroups in DMIST. Radiology 2008; 246: pp. 376-383.
6. Bloomquist A.K., Yaffe M.J., Pisano E.D., et. al.: Quality control for digital mammography in the ACRIN DMIST trial: part I. Med Phys 2006; 33: pp. 719-736.
7. Yaffe M.J., Bloomquist A.K., Mawdsley G.E., et. al.: Quality control for digital mammography: part II. Recommendations from the ACRIN DMIST trial. Med Phys 2006; 33: pp. 737-752.
8. Irving B.J., Maree G.J., Hering E.R., et. al.: Radiation dose from a linear slit scanning X-ray machine with full-body imaging capabilities. Radiat Prot Dosimetry 2008; 130: pp. 482-489.
9. Hauge I.H., Pedersen K., Sanderud A., et. al.: Patient doses from screen-film and full-field digital mammography in a population-based screening programme. Radiat Prot Dosimetry 2012; 148: pp. 65-73.
10. Aslund M., Cederstrom B., Lundqvist M., et. al.: Scatter rejection in multislit digital mammography. Med Phys 2006; 33: pp. 933-940.
11. McCullagh J.B., Baldelli P., Phelan N.: Clinical dose performance of full field digital mammography in a breast screening programme. Br J Radiol 2011; 84: pp. 1027-1033.
12. Perry N., Broeders M.J., de Wolf C.: European Guidelines for Quality Assurance in Breast Cancer Screening and Diagnosis.2006.Office for Official Publications of the European CommunitiesLuxembourg
13. Dance D.R., Young K.C., van Engen R.E.: Further factors for the estimation of mean glandular dose using the United Kingdom, European and IAEA breast dosimetry protocols. Phys Med Biol 2009; 54: pp. 4361-4372.
14. D’Orsi C.J., Bassett L.W., Berg W.A., et. al.: BI-RADS: mammography.D’Orsi C.J.Mendelson E.B.Ikeda D.M. et. al.Breast Imaging Reporting and Data System: ACR BI-RADS—Breast Imaging Atlas.2003.American College of RadiologyReston, VA:
15. Dorfman D., Alf E.: Maximum likelihood estimation of parameters of signal detection theory and determination of confidence intervals—rating method data. J Math Psychol 1969; 6: pp. 487-496.
16. Metz C., Pan X.: “Proper” binormal ROC curves: theory and maximum-likelihood estimation. J Math Psychol 1999; 43: pp. 1-33.
17. Pesce L., Metz C.: Reliable and computationally efficient maximum-likelihood estimation of “proper” binormal ROC curves. Acad Radiol 2007; 14: pp. 814-829.
18. Dorfman D.D., Berbaum K.S., Metz C.E.: Receiver operating characteristic rating analysis: generalization to the population of readers and patients with the jackknife method. Invest Radiol 1992; 27: pp. 723-731.
19. Chakraborty D., Berbaum K.S.: Observer studies involving detection and localization: modeling, analysis, and validation. Med Phys 2004; 31: pp. 2313-2330.
20. Chakraborty D. JAFROC software. Available at: http://www.devchakraborty.com/JAFROC.html . Accessed October 12, 2011.
21. Sidik K.: Exact unconditional tests for testing non-inferiority in matched-pairs design. Stat Med 2003; 22: pp. 265-278.
22. Obuchowski N.A., Rockette H.E.: Hypothesis testing of diagnostic accuracy for multiple readers and multiple tests: an ANOVA approach with dependent observations. Commun Stat B Simul Comput 1995; 24: pp. 285-308.
23. Hillis S.L.: A comparison of denominator degrees of freedom methods for multiple observer ROC analysis. Stat Med 2007; 26: pp. 596-619.
24. Cole E., Pisano E.D., Brown M., et. al.: Diagnostic accuracy of Fischer Senoscan Digital Mammography versus screen-film mammography in a diagnostic mammography population. Acad Radiol 2004; 11: pp. 879-886.
25. Sectra Imtec AB. Clinical Data Report Sectra MicroDose Mammography. Linköping, Sweden: Sectra Imtec AB; 2010. p. DOC-EMOA-7D5AT2-2.0.
26. Pisano E.D., Acharyya S., Cole E.B., et. al.: Cancer cases from ACRIN digital mammographic imaging screening trial: radiologist analysis with use of a logistic regression model. Radiology 2009; 252: pp. 348-357.