RESUMO
OBJECTIVE: To assess breast imaging findings, biopsy rates, and malignancy rates in areas of palpable concern in women at high risk for breast cancer. METHODS: An IRB-approved retrospective review of a tertiary cancer center's breast imaging database was performed. Breast imaging and electronic medical records of high-risk women with palpable findings detected on self- or clinical breast examination from January 1, 2010, to January 1, 2016, were reviewed. Descriptive statistical analyses were conducted. RESULTS: Imaging correlates for 322 palpable findings in 238 high-risk women included 55/203 (27.1%) on mammography, 183/302 (60.6%) on US, and 20/47 (42.6%) on MRI. Biopsies were performed for 104/322 (32.3%) palpable findings: 95/104 (91.3%) under imaging guidance and 9/104 (8.7%) under palpation after negative imaging. Of 322 palpable findings, 16 (5.0%) were malignant in 16/238 (6.7%) women, yielding a positive predictive value of biopsy of 16.8% (95% CI: 9.2%-24%). Women diagnosed with cancer had 16/16 (100%) sonographic, 9/14 (64.3%) mammographic, and 7/7 (100%) MRI correlates. Cancer histopathology included 12 invasive ductal carcinomas, 1 ductal carcinoma in situ, 1 invasive lobular carcinoma, 1 malignant phyllodes tumor, and 1 metastatic carcinoid tumor. Over two years of follow-up imaging in 183/238 (76.9%) women were reviewed; 7/183 (3.8%) were diagnosed with breast cancer at least one year after presenting with a palpable concern in a different location. CONCLUSION: High-risk women with palpable findings exhibit a 6.7% malignancy rate, indicating the value of imaging workup in this population. In our cohort, imaging demonstrated a high negative predictive value.
RESUMO
OBJECTIVE: To investigate the feasibility of using deep learning to identify tumor-containing axial slices on breast MRI images. METHODS: This IRB-approved retrospective study included consecutive patients with operable invasive breast cancer undergoing pretreatment breast MRI between January 1, 2014, and December 31, 2017. Axial tumor-containing slices from the first postcontrast phase were extracted. Each axial image was subdivided into two subimages: one of the ipsilateral cancer-containing breast and one of the contralateral healthy breast. Cases were randomly divided into training, validation, and testing sets. A convolutional neural network was trained to classify subimages into "cancer" and "no cancer" categories. Accuracy, sensitivity, and specificity of the classification system were determined using pathology as the reference standard. A two-reader study was performed to measure the time savings of the deep learning algorithm using descriptive statistics. RESULTS: Two hundred and seventy-three patients with unilateral breast cancer met study criteria. On the held-out test set, accuracy of the deep learning system for tumor detection was 92.8% (648/706; 95% confidence interval: 89.7%-93.8%). Sensitivity and specificity were 89.5% and 94.3%, respectively. Readers spent 3 to 45 seconds to scroll to the tumor-containing slices without use of the deep learning algorithm. CONCLUSION: In breast MR exams containing breast cancer, deep learning can be used to identify the tumor-containing slices. This technology may be integrated into the picture archiving and communication system to bypass scrolling when viewing stacked images, which can be helpful during nonsystematic image viewing, such as during interdisciplinary tumor board meetings.