Improved cancer detection in automated breast ultrasound by radiologists using Computer Aided Detection.

OBJECTIVE: To investigate the effect of dedicated Computer Aided Detection (CAD) software for automated breast ultrasound (ABUS) on the performance of radiologists screening for breast cancer. METHODS: 90 ABUS views of 90 patients were randomly selected from a multi-institutional archive of cases collected between 2010 and 2013. This dataset included normal cases (n=40) with >1year of follow up, benign (n=30) lesions that were either biopsied or remained stable, and malignant lesions (n=20). Six readers evaluated all cases with and without CAD in two sessions. CAD-software included conventional CAD-marks and an intelligent minimum intensity projection of the breast tissue. Readers reported using a likelihood-of-malignancy scale from 0 to 100. Alternative free-response ROC analysis was used to measure the performance. RESULTS: Without CAD, the average area-under-the-curve (AUC) of the readers was 0.77 and significantly improved with CAD to 0.84 (p=0.001). Sensitivity of all readers improved (range 5.2-10.6%) by using CAD but specificity decreased in four out of six readers (range 1.4-5.7%). No significant difference was observed in the AUC between experienced radiologists and residents both with and without CAD. CONCLUSIONS: Dedicated CAD-software for ABUS has the potential to improve the cancer detection rates of radiologists screening for breast cancer.

3D quantitative breast ultrasound analysis for differentiating fibroadenomas and carcinomas smaller than 1cm.

PURPOSE: In (3D) ultrasound, accurate discrimination of small solid masses is difficult, resulting in a high frequency of biopsies for benign lesions. In this study, we investigate whether 3D quantitative breast ultrasound (3DQBUS) analysis can be used for improving non-invasive discrimination between benign and malignant lesions. METHODS AND MATERIALS: 3D US studies of 112 biopsied solid breast lesions (size <1cm), were included (34 fibroadenomas and 78 invasive ductal carcinomas). The lesions were manually delineated and, based on sonographic criteria used by radiologists, 3 regions of interest were defined in 3D for analysis: ROI (ellipsoid covering the inside of the lesion), PER (peritumoural surrounding: 0.5mm around the lesion), and POS (posterior-tumoural acoustic phenomena: region below the lesion with the same size as delineated for the lesion). After automatic gain correction (AGC), the mean and standard deviation of the echo level within the regions were calculated. For the ROI and POS also the residual attenuation coefficient was estimated in decibel per cm [dB/cm]. The resulting eight features were used for classification of the lesions by a logistic regression analysis. The classification accuracy was evaluated by leave-one-out cross-validation. Receiver operating characteristic (ROC) curves were constructed to assess the performance of the classification. All lesions were delineated by two readers and results were compared to assess the effect of the manual delineation. RESULTS: The area under the ROC curve was 0.86 for both readers. At 100% sensitivity, a specificity of 26% and 50% was achieved for reader 1 and 2, respectively. Inter-reader variability in lesion delineation was marginal and did not affect the accuracy of the technique. The area under the ROC curve of 0.86 was reached for the second reader when the results of the first reader were used as training set yielding a sensitivity of 100% and a specificity of 40%. Consequently, 3DQBUS would have achieved a 40% reduction in biopsies for benign lesions for reader 2, without a decrease in sensitivity. CONCLUSION: This study shows that 3DQBUS is a promising technique to classify suspicious breast lesions as benign, potentially preventing unnecessary biopsies.