Original investigationImproved Differential Diagnosis of Breast Masses on Ultrasonographic Images with a Computer-Aided Diagnosis Scheme for Determining Histological Classifications
Section snippets
Materials and methods
Institutional review board approval was obtained for this study.
Results
For the 100 cases used in the observer study, the sensitivity, the specificity, and the AUC of the CAD scheme trained by use of the other cases were 90% (45/50), 92% (46/50), and 0.95, respectively.
Figure 2 shows the average ROC curves for all observers in distinguishing between benign and malignant masses without and with CAD. The average AUC for all observers increased from 0.716 without to 0.864 with CAD (P = .006). Table 1 shows the AUCs for each observer with and without CAD. All
Discussion
The observer study showed that the AUCs for all observers increased with the likelihood of histological classification evaluated by the CAD scheme. The gain in the average AUCs for the general group was higher than that for the expert group. The general clinicians probably tended to accept the computer output comparatively easily. This would imply that the performance of the CAD scheme is very important. Although the average AUCs for the expert group were also improved with the CAD scheme, the
Conclusion
The presentation of the likelihood of the histological classification evaluated by the CAD scheme showed beneficial effects for most cases, and it improved the clinicians' performance in the observer study. The CAD scheme is therefore considered to be useful for clinicians in making a differential diagnosis of masses on ultrasonographic images.
References (14)
- et al.
Using sonography to screen women with mammographically dense breasts
Am J Roentegenol AJR
(2003) - et al.
Ultrasound screening of breast cancer
Breast Cancer
(2009) - et al.
Combined screening with ultrasound and mammography vs mammography alone in women at elevated risk of breast cancer
JAMA
(2008) - et al.
A comparison of mammography and ultrasound in women with breast disease: a receiver operating characteristic analysis
Breast J
(2012) - et al.
Occurrence of focal detection abnormalities in breast cancer screening by ultrasound
J Jpn Assoc Breast Cancer Screen
(2000) Computer-aided diagnosis in medical imaging: historical review, current status and future potential
Comput Med Imaging Graph
(2007)- et al.
Improving breast cancer diagnosis with computer-aided diagnosis
Acad Radiol
(1999)