Rationale and Objectives
To compare time to interpretation and diagnostic performance levels during repeat readings of full-field digital mammography (FFDM) and digital breast tomosynthesis (DBT) in a retrospective study.
Materials and Methods
Three experienced radiologists twice interpreted 125 selected examinations, 35 with verified cancers and 90 negative for cancer during a period of 22 months using FFDM alone followed by a combined FFDM + DBT mode. Changes in time to “review and rate” these examinations as well as in diagnostic performance levels where assessed. A fixed-effect analysis accounting for cross-correlation due to the review of the same examinations by the same readers was performed.
Results
The total (combined) time to review and rate an examination increased on average by 33% between the first and second readings of the same examinations ( P < .001). Radiologists reduced their time to review FFDM before making the DBT available for viewing. However, they spent more time reviewing the combined FFDM + DBT mode. The recall rates for examinations depicting cancer remained largely unchanged. Among the groups of examinations with concordant and discordant recall recommendations during the two readings only the group examinations that were “newly recalled” during repeat reading, took significantly longer ( P < .01).
Conclusion
DBT-based breast imaging may ultimately result in a substantial increase in performance; however, without efficiency improvements DBT may take longer to interpret. Addition of “false-positive recalls” was most strongly associated with increase in interpretation time while elimination of “false-positive recalls” did not require longer interpretation time.
With the anticipated availability of digital breast tomosynthesis (DBT) imaging systems for three-dimensional breast examinations , there are a number of questions related to acquisition, operation, training, and display that need to be addressed if DBT is to be optimally incorporated into the clinical practice. DBT is of great interest in screening, as well as in diagnostic procedures since it enables three-dimensional reconstruction, thus allowing cross-sectional visualization of breast tissue. This increase in visualization reduces the difficulty associated with interpretations of projection mammograms because of superposition or “overlapping” tissue. Several studies have addressed technical, ergonomic, and performance issues associated with DBT, but to date there are no data from prospective clinical studies. The results from retrospective or a number of primarily subjective assessments are quite encouraging but inconclusive . In two retrospective studies performed in our laboratory, the time to interpret examinations using either DBT alone or full-field digital mammography (FFDM) + DBT combined was longer than when interpreting FFDM . Because the issue of “insufficient training and familiarity with DBT” was raised as one possible reason for substantially longer review times with DBT-based systems, in particular as related to the possible use of DBT in the screening environment where efficiency is of great importance, we performed a repeat reading study in which three experienced radiologists read the same set of 125 examinations twice under the same reading mode. We present here the results of this very preliminary study, albeit potentially important in terms of highlighting an issue that we feel needs to be better understood and adequately addressed.
Materials and methods
General Study Design
The digital mammography images used in this study, consisting of FFDM images and DBT image sets, have been described in detail elsewhere . In brief, images were acquired with standard tomosynthesis acquisition techniques, which generated 11 low-dose projection images, or frames, acquired for reconstruction of DBT image sets. After acquisition, the data from the frames were used to reconstruct 50–90 parallel slices 1 mm thick, depending on the thickness of the compressed breast. The radiation dose associated with the series of low-dose projection images was the same as that of a projection mammogram with average mid-breast dose of approximately 2 mGy per view. The acquisition of all examinations was performed under Institutional Review Board–approved protocols that included a signed informed consent by the participant. The reading study used an FFDM alone mode and a “combined” mode (FFDM + DBT).
Get Radiology Tree app to read full this article<
Breast Findings
Get Radiology Tree app to read full this article<
Table 1
Distribution of Examinations with Single versus Multiple Abnormalities
Mode FFDM DBT Abnormality Total Positive Total Positive Single Mass 39 21 (53.8%) 39 19 (48.7%) Calcifications 10 2 (20.0%) 9 2 (22.2%) 49 23 (46.9%) 48 21 (43.8%) Multiple Two or more masses 4 1 (25.0%) 8 3 (37.5%) Two or more calcifications – – – – Both ∗ 14 11 (78.6%) 12 11 (91.7%) 18 12 (66.7%) 20 14 (70.0%)
DBT, digital breast tomosynthesis; FFDM, full-field digital mammography.
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Observers
Get Radiology Tree app to read full this article<
Performance of the Study
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Data Analysis
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<
Results
Get Radiology Tree app to read full this article<
Table 2
Median Interpretation-and-rating Times (minutes) during the Assessments of 90 Cancer-free Examinations and 35 Examinations Depicting Verified Cancers for FFDM Alone and FFDM + DBT Modes
FFDM Alone FFDM + DBT Combo Entire Sequence (Total Time ‡ ) Type of Examinations Readers Initial Reading Repeat Reading Initial Reading Repeat Reading Initial Reading Repeat Reading
DBT, digital breast tomosynthesis; FFDM, full-field digital mammography.
∗ Statistically significant (α = 0.05) changes in interpretation time between initial and repeat readings.
† Framed cells correspond to groups in which changes in interpretation time on average was statistically significant (α = 0.05). All significant P values are P < .001.
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
Recall Rates for the 90 Examinations not Depicting Cancer (“False-positive” Rates) under the Combined FFDM + DBT Mode
Reader Initial Reading Repeat Reading Difference 1 0.34 0.43 0.09 ∗ 2 0.73 0.58 −0.16 ∗ 3 0.38 0.28 −0.10 ∗ Average 0.49 0.43 −0.06
DBT, digital breast tomosynthesis; FFDM, full-field digital mammography.
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<
Discussion
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Conclusion
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
References
1. Rafferty E.A.: Digital mammography: novel applications. Radiol Clin North Am 2007; 45: pp. 831-843.
2. Lewin J.M., Niklason L.: Advanced applications of digital mammography: tomosynthesis and contrast-enhanced digital mammography. Semin Roentgenol 2007; 42: pp. 243-252.
3. Niklason L.T., Christian B.T., Niklason L.E., et. al.: Digital tomosynthesis in breast imaging. Radiology 1997; 205: pp. 399-406.
4. Niklason L.T., Kopans D.B., Hamberg L.M.: Digital breast imaging: tomosynthesis and digital subtraction mammography. Breast Dis 1998; 10: pp. 151-164.
5. Suryanarayanan S., Karellas A., Vedantham S., et. al.: Evaluation of linear and non-linear tomosynthetic reconstruction methods in digital mammography. Acad Radiol 2001; 8: pp. 219-224.
6. Wu T., Moore R.H., Rafferty E.A., et. al.: A comparison of reconstruction algorithms for breast tomosynthesis. Med Phys 2004; 31: pp. 2636-2647.
7. Smith A.: Full-field breast tomosynthesis. Radiol Manage 2005; 27: pp. 25-31.
8. Andersson I., Ikeda D.M., Zackrisson S., et. al.: Breast tomosynthesis and digital mammography: a comparison of breast cancer visibility and BIRADS classification in a population of cancers with subtle mammographic findings. Eur Radiol 2008; 18: pp. 2817-2825.
9. Moore R.H., Stewart A., Wu T., et. al.: Second-generation digital breast tomosynthesis (DBT) in the screening setting: workflow and preliminary results. Presented at RSNA 2005; [SSG01–SSG07]
10. Chan H.P., Wei J., Sahiner B., et. al.: Computer-aided detection system for breast masses on digital tomosynthesis mammograms: preliminary experience. Radiology 2005; 237: 1075–1018
11. Chen S.C., Carton A.K., Albert M., et. al.: Initial clinical experience with contrast-enhanced digital breast tomosynthesis. Acad Radiol 2007; 14: pp. 229-238.
12. Poplack S.P., Tosteson T.D., Kogel C.H., et. al.: Digital breast tomosynthesis: initial experience in 98 women with abnormal digital screening mammography. AJR Am J Roentgenol 2007; 189: pp. 616-623.
13. Good W.F., Abrams G.S., Catullo V.J., et. al.: Digital breast tomosynthesis: a pilot observer study. AJR Am J Roentgenol 2008; 190: pp. 865-869.
14. Gur D., Abrams G.S., Chough D.M., et. al.: Digital breast tomosynthesis—an observer performance study. AJR Am J Roentgenol 2009; 193: pp. 586-591.
15. Gur D., Rockette H.E.: Performance assessments of diagnostic systems under the FROC paradigm: experimental, analytical, and results interpretation issues. Acad Radiol 2008; 15: pp. 1312-1315.
16. Reiser I., Nishikawa R.M., Giger M.L., Wu , et. al.: Computerized mass detection for digital breast tomosynthesis directly from the projection images. Med Phys 2006; 33: pp. 482-491.
17. Reiser I., Nishikawa R.M., Giger M.L., et. al.: Computerized detection of mass lesions in digital breast tomosynthesis images using two- and three dimensional radial gradient index segmentation. Technol Cancer Res Treat 2004; 3: pp. 437-441.
18. Reiser I., Nishikawa R.M., Edwards A.V., et. al.: Automated detection of microcalcification clusters for digital breast tomosynthesis using projection data only: a preliminary study. Med Phys 2008; 35: pp. 1486-1493.
19. Moore R.H., Levy A., Stewart A., et. al.: Reading behavior for screening digital breast tomosynthesis (DBT) compared to conventional 2D mammography (CM). Presented at RSNA 2006; [SSG01–SSG05]
20. Kundel H.L., Nodine C.F., Conant E.F., et. al.: Holistic component of image perception in mammogram interpretation: gaze-tracking study. Radiology 2007; 242: pp. 396-402.
21. Mello-Thoms C., Hardesty L., Sumkin J., et. al.: Effects of lesion conspicuity on visual search in mammogram reading. Acad Radiol 2005; 12: pp. 830-840.