A comparative analysis of 2D and 3D CAD for calcifications in digital breast tomosynthesis

Abstract

Many medical centers offer digital breast tomosynthesis (DBT) and 2D digital mammography acquired under the same compression (i.e., "Combo" examination) for screening. This paper compares a conventional 2D CAD algorithm (Hologic ® ImageChecker ® CAD v9.4) for calcification detection against a prototype 3D algorithm (Hologic ® ImageChecker ® 3D Calc CAD v1.0). Due to the newness of DBT, the development of this 3D CAD algorithm is ongoing, and it is currently not FDA-approved in the United States. For this study, DBT screening cases with suspicious calcifications were identified retrospectively at the University of Pennsylvania. An expert radiologist (E.F.C.) reviewed images with both 2D and DBT CAD marks, and compared the marks to biopsy results. Control cases with one-year negative follow-up were also studied; these cases either possess clearly benign calcifications or lacked calcifications. To allow the user to alter the sensitivity for cancer detection, an operating point is assigned to each CAD mark. As expected from conventional 2D CAD, increasing the operating point in 3D CAD increases sensitivity and reduces specificity. Additionally, we showed that some cancers are occult to 2D CAD at all operating points. By contrast, 3D CAD allows for detection of some cancers that are missed on 2D CAD. We also demonstrated that some non-cancerous CAD marks in 3D are not present at analogous locations in the 2D image. Hence, there are additional marks when using both 2D and 3D CAD in combination, leading to lower specificity than with conventional 2D CAD alone.

Multi‐domain features for reducing false positives in automated detection of clustered microcalcifications in digital breast tomosynthesis

“…While MC detection methods developed in 2D mammograms may be translated to the projection views or tomographic slices, 2D methods cannot take full advantage of 3D spatial contextual information of DBT volume for MC detection. A few 3D approaches have been proposed for MC detection in DBT volume, yet the performance of these approaches need to be improved . A MC cluster split into different slices may be more easily recognized on planar projection image, as the scattered MCs are gathered into cluster by maximum intensity projection.…”

“…High false positives (FPs) is a key issue in MC detection from DBT volume, where some FP detections reveal similar morphological appearance to true MCs. Vessels and fibrous structures may extend across slices and be detected as separated objects due to noise or blurring artifacts in adjacent slices . In order to maintain high sensitivity for MC detection, the detection algorithms usually mark more MC candidates that may include many FPs.…”

“…A few 3D approaches [16][17][18][19] have been proposed for MC detection in DBT volume, yet the performance of these approaches need to be improved. 10,20 A MC cluster split into different slices may be more easily recognized on planar projection image, 21 as the scattered MCs are gathered into cluster by maximum intensity projection. Blurring arises from the quasi-3D DBT volume due to reconstruction from multiple limited-angle projections and the focal spot blur resulting from continuous x-ray source motion.…”

“…Vessels and fibrous structures may extend across slices and be detected as separated objects due to noise 18 or blurring artifacts in adjacent slices. 20 In order to maintain high sensitivity for MC detection, the detection algorithms usually mark more MC candidates that may include many FPs. Therefore, reducing FPs in MC detection represents a true challenge.…”