Computer-aided detection system for masses in automated whole breast ultrasonography: development and evaluation of the effectiveness

PURPOSE: The aim of this study was to evaluate the performance of a proposed computer-aided detection (CAD) system in automated breast ultrasonography (ABUS). METHODS: Eighty-nine two-dimensional images (20 cysts, 42 benign lesions, and 27 malignant lesions) were obtained from 47 patients who underwent ABUS (ACUSON S2000). After boundary detection and removal, we detected mass candidates by using the proposed adjusted Otsu's threshold; the threshold was adaptive to the variations of pixel intensities in an image. Then, the detected candidates were segmented. Features of the segmented objects were extracted and used for training/testing in the classification. In our study, a support vector machine classifier was adopted. Eighteen features were used to determine whether the candidates were true lesions or not. A five-fold cross validation was repeated 20 times for the performance evaluation. The sensitivity and the false positive rate per image were calculated, and the classification accuracy was evaluated for each feature. RESULTS: In the classification step, the sensitivity of the proposed CAD system was 82.67% (SD, 0.02%). The false positive rate was 0.26 per image. In the detection/segmentation step, the sensitivities for benign and malignant mass detection were 90.47% (38/42) and 92.59% (25/27), respectively. In the five-fold cross-validation, the standard deviation of pixel intensities for the mass candidates was the most frequently selected feature, followed by the vertical position of the centroids. In the univariate analysis, each feature had 50% or higher accuracy. CONCLUSION: The proposed CAD system can be used for lesion detection in ABUS and may be useful in improving the screening efficiency.

A Method to Quantify Breast MRI for Predicting Tumor Invasion in Patients with Preoperative Biopsy- Proven Ductal Carcinoma in Situ (DCIS)

PURPOSE: To determine the quantitative parameters of breast MRI that predict tumor invasion in biopsy-proven DCIS. MATERIALS AND METHODS: From January 2009 to March 2010, 42 MRI examinations of 41 patients with biopsy-proven DCIS were included. The quantitative parameters, which include the initial percentage enhancement (E1), peak percentage enhancement (E(peak)), time to peak enhancement (TTP), signal enhancement ratio (SER), arterial enhancement fraction (AEF), apparent diffusion coefficient (ADC) value, long diameter and the volume of the lesion, were calculated as parameters that might predict invasion. Univariate and multivariate analyses were used to identify the parameters associated with invasion. RESULTS: Out of 42 lesions, 23 lesions were confirmed to be invasive ductal carcinoma (IDC) and 19 lesions were confirmed to be pure DCIS. Tumor size (p = 0.003; 6.5 +/- 3.2 cm vs. 3.6 +/- 2.6 cm, respectively) and SER (p = 0.036; 1.1 +/- 0.3 vs. 0.9 +/- 0.3, respectively) showed statistically significant high in IDC. In contrast, E1, Epeak, TTP, ADC, AEF and volume of the lesion were not statistically significant. Tumor size and SER had statistically significant associations with invasion, with an odds ratio of 1.04 and 22.93, respectively. CONCLUSION: Of quantitative parameters analyzed, SER and the long diameter of the lesion could be specific parameter for predicting invasion in the biopsy-proven DCIS.

Dose Reduction in Automatic Optimization Parameter of Full Field Digital Mammography: Breast Phantom Study / 한국유방암학회지

PURPOSE: We evaluated the impact of three automatic optimization of parameters (AOP) modes of digital mammography on the dose and image quality. METHODS: Computerized Imaging Reference Systems phantoms were used. A total of 12 phantoms with different thickness and glandularity were imaged. We analyzed the average glandular dose (AGD) and entrance surface exposure (ESE) of 12 phantoms imaged by digital mammography in three modes of AOP; namely standard mode (STD), contrast mode (CNT), and dose mode (DOSE). Moreover, exposure factors including kVp, mAs, and target/filter combination were evaluated. To evaluate the quality of the obtained digital image, two radiologists independently counted the objects of the phantoms. RESULTS: According to the AOP modes, the score of masses and specks was sorted as CNT>STD=DOSE. There was no difference in the score of fiber among the three modes. The score of image preference was sorted as CNT>STD>DOSE. The AGD, ESE, and mAs were sorted as CNT>STD>DOSE. The kVp was sorted as CNT=STD>DOSE. The score of all test objects in the phantom image was on a downtrend with increasing breast thickness. The score of masses was different among the three groups; 20-21%>30%>50% glandularity. The score of specks was sorted as 20-21%=30%>50% glandularity. The score of fibers was sorted as 30%>20-21%=50% glandularity. The score of image preference was not different among the three glandularity groups. The AGD, ESE, kVp, and mAs were correlated with breast thickness, but not correlated with glandularity. CONCLUSION: The DOSE mode offers significant improvement (19.1-50%) in dose over the other two modes over a range of breast thickness and breast glandularity with acceptable image quality. Owning knowledge of the three AOP modes may reduce unnecessary radiation exposure by utilizing the proper mode according to its purpose.

A Rare Case of Recurrent Myoid Hamartoma Mimicking Malignancy: Imaging Appearances

Myoid hamartoma is an uncommon type of breast hamartoma and its recurrence is very rare. We report the imaging appearance of an unusual case of recurrent myoid hamartoma of the breast mimicking malignancy in a 43-year-old woman. Although the mammographic and ultrasonographic findings have long been described in the literature, MR finding with a dynamic study has not, to the best of our knowledge, been reported previously.

Painless Posterior Thigh Mass as a Presentation of Metastatic Breast Cancer / 한국유방암학회지

Soft tissue metastasis clinically presenting as either painless subcutaneous or painful intramuscular nodules is extremely rare and may lead to an errant clinical suspicion of sarcoma. In general, most soft tissue metastasis comes from lung carcinoma; however, to date, there have been no reports of a posterior thigh mass just beneath the skin metastasizing from breast cancer. Here, we report a case of distant soft tissue metastasis presenting as a painless solitary left posterior thigh mass measuring 1.5 cm in diameter, which was later shown by positron emission tomography-computed tomography (PET-CT) to have multiple simultaneous mediastinal lymph node metastases. Eleven months ago, the patient had undergone curative breast-conserving surgery (BCS) for cancer of her left breast. At that time, her tumor showed a triple negative phenotype. Initial PET-CT right before the BCS had shown no metastasis. After histological and radiologic evaluation for the metastases, she decided to have systemic chemotherapy.

Diffusion-Weighted MR Images for Hyperacute Cerebral Infarction: Design of a Quick Volume Estimation Method for Hyperintensities

PURPOSE: To design a reliable and quick lesion volume estimation method for hyperintensities on diffusion-weighted images (DWI) for the evaluation of hyperacute stroke. MATERIALS AND METHODS: Twenty patients with obvious high signal lesions seen on DWI in the middle cerebral artery territory due to acute ischemia were enrolled to evaluate the performance of four tentatively designed semi-quantitative methods: the 25-area method, the 20-area method, the 10-area method, and the modified 10-area method. Two radiologists performed the volume analyses using these methods. Intraclass correlation coefficients were calculated to compare the correlation between the reference values and the measured values and to evaluate the interobserver agreement of each method. RESULTS: For the correlation between the measured value and the reference value, the performance of the modified 10-area method was the most powerful, with a value of 0.8981 and 0.8090 for observer 1 and 2, respectively. The interobserver agreement was satisfactory for both the 25-area method and the modified 10-area method, with a value of 0.9212 (95% CI: 0.8123-0.9681) and 0.9063 (95% CI: 0.7790-0.9618), respectively. CONCLUSION: The performance of the modified 10-area method was satisfactory for both lesion volume estimation and interobserver correlation in the evaluation of an acute cerebral infarction by the use of DWI.