Prediction of malignancy by a radiomic signature from contrast agent‐free diffusion MRI in suspicious breast lesions found on screening mammography.

171

155

mentioning

confidence: 99%

Preoperative prediction of sentinel lymph node metastasis in breast cancer by radiomic signatures from dynamic contrast‐enhanced MRI

Liu

Ding

Spuhler

et al. 2018

171

155

“…Compared with traditional methods, the main advantage of deep learning models is its capability in extracting highly representative features in a data‐driven way. Recently, the efficacy of deep neural networks has been evaluated in breast cancer classification tasks . However, these works either used a small‐size dataset or needed manual annotations on lesions during the training phase.…”

Section: Discussionmentioning

confidence: 99%

“…Recently, the efficacy of deep neural networks has been evaluated in breast cancer classification tasks. [28][29][30] However, these works either used a small-size dataset or needed manual annotations on lesions during the training phase. Directly localizing breast cancers in 3D radiology images with only image-level supervision has not yet been extensively explored.…”

Section: Discussionmentioning

confidence: 99%

Weakly supervised 3D deep learning for breast cancer classification and localization of the lesions in MR images

Zhou

Luo

Dou

et al. 2019

119

Background The usefulness of 3D deep learning‐based classification of breast cancer and malignancy localization from MRI has been reported. This work can potentially be very useful in the clinical domain and aid radiologists in breast cancer diagnosis. Purpose To evaluate the efficacy of 3D deep convolutional neural network (CNN) for diagnosing breast cancer and localizing the lesions at dynamic contrast enhanced (DCE) MRI data in a weakly supervised manner. Study Type Retrospective study. Subjects A total of 1537 female study cases (mean age 47.5 years ±11.8) were collected from March 2013 to December 2016. All the cases had labels of the pathology results as well as BI‐RADS categories assessed by radiologists. Field Strength/Sequence 1.5 T dynamic contrast‐enhanced MRI. Assessment Deep 3D densely connected networks were trained under image‐level supervision to automatically classify the images and localize the lesions. The dataset was randomly divided into training (1073), validation (157), and testing (307) subsets. Statistical Tests Accuracy, sensitivity, specificity, area under receiver operating characteristic curve (ROC), and the McNemar test for breast cancer classification. Dice similarity for breast cancer localization. Results The final algorithm performance for breast cancer diagnosis showed 83.7% (257 out of 307) accuracy (95% confidence interval [CI]: 79.1%, 87.4%), 90.8% (187 out of 206) sensitivity (95% CI: 80.6%, 94.1%), 69.3% (70 out of 101) specificity (95% CI: 59.7%, 77.5%), with the area under the curve ROC of 0.859. The weakly supervised cancer detection showed an overall Dice distance of 0.501 ± 0.274. Data Conclusion 3D CNNs demonstrated high accuracy for diagnosing breast cancer. The weakly supervised learning method showed promise for localizing lesions in volumetric radiology images with only image‐level labels. Level of Evidence: 4 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2019;50:1144–1151.

“…Computer‐assisted diagnosis software was introduced more than a decade ago, but recent developments using neuronal networks and multivariate feature extraction ("radiomics") might have a substantial impact in supporting the image evaluation of radiologists in abbreviated breast imaging protocols: Using such software algorithms, which are known to be able to extract up to 40,000 different image characteristics, might increase diagnostic accuracy, reproducibility, and clinical efficacy of abbreviated protocols. Further, it has been demonstrated that such analyses might even give further insight in histological subtypes and provide prognostic information .…”

Section: Emerging Techniques and Outlookmentioning

confidence: 99%

Abbreviated MRI Protocols in Breast Cancer Diagnostics

Deike-Hofmann

Koenig

Paech

et al. 2018

Self Cite

Oncologic imaging focused on the detection of breast cancer is of increasing importance, with over 1.7 million new cases detected each year worldwide. MRI of the breast has been described to be one of the most sensitive imaging modalities in breast cancer detection; however, clinical use is limited due to high costs. In the past, the objective and clinical routine of oncologic imaging was to provide one extended imaging protocol covering all potential needs and clinical implications regardless of the specific clinical indication or question. Future protocols might be more focused according to a "keep it short and simple" approach, with a reduction of patient magnet time and a limited number of images to review. Rather than replacing conventional full‐diagnostic breast MRI protocols, these approaches aim at introducing a new thinking in oncologic imaging using a diversification of available imaging approaches targeted to the dedicated clinical needs of the individual patient. Here we review current approaches on using abbreviated protocols that aim to increase the clinical availability and use of breast MRI for improved early detection of breast cancer. Level of Evidence: 2 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2019;49:647–658.