Automated Breast Ultrasound Lesions Detection Using Convolutional Neural Networks

Abstract

Breast lesion detection using ultrasound imaging is considered an important step of computer-aided diagnosis systems. Over the past decade, researchers have demonstrated the possibilities to automate the initial lesion detection. However, the lack of a common dataset impedes research when comparing the performance of such algorithms. This paper proposes the use of deep learning approaches for breast ultrasound lesion detection and investigates three different methods: a Patch-based LeNet, a U-Net, and a transfer learning approach with a pretrained FCN-AlexNet. Their performance is compared against four state-of-the-art lesion detection algorithms (i.e., Radial Gradient Index, Multifractal Filtering, Rule-based Region Ranking, and Deformable Part Models). In addition, this paper compares and contrasts two conventional ultrasound image datasets acquired from two different ultrasound systems. Dataset A comprises 306 (60 malignant and 246 benign) images and Dataset B comprises 163 (53 malignant and 110 benign) images. To overcome the lack of public datasets in this domain, Dataset B will be made available for research purposes. The results demonstrate an overall improvement by the deep learning approaches when assessed on both datasets in terms of True Positive Fraction, False Positives per image, and F-measure.

Breast mass classification in sonography with transfer learning using a deep convolutional neural network and color conversion

“…The OASBUD was originally used to assess the statistical properties of backscattered ultrasound echoes in breast tissue and to differentiate breast masses using transfer learning with CNNs . Detailed descriptions of both datasets can be found in the original papers …”

“…To show the usefulness of the methods proposed in this paper, we also employed two publicly available breast mass datasets. 26,32 The first one, named UDIAT, consists of 163 B-mode images of breast masses (53 malignant and 110 benign) collected using Siemens ACUSON scanner from the UDIAT Diagnostic Centre of the Parc Tauli Corporation, Sabadell (Spain). This dataset was used by the authors to develop deep learning-based algorithms for the breast mass detection 26 and segmentation.…”

“…Transfer learning methods were employed for breast mass classification and segmentation in several studies. [19][20][21][22][23][24][25] Additionally, deep learning was used to detect breast lesions 26 and differentiate breast masses with shear-wave elastography. 27 The betterperforming pretrained deep learning models have been developed using RGB color images.…”

Artificial intelligence in cancer imaging: Clinical challenges and applications

“…Although CADe continues to be developed for screening mammography, investigators also have sought to automate the detection of breast lesions on 3D ultrasound, breast MRI, and breast tomosynthesis images by incorporating predefined algorithms as well as novel deep learning methods . The motivation for computerized detection on 3D breast images arose with the arrival of 3D ultrasound and MRI for use as adjunct imaging for screening women with dense breast tissue …”

“…CNNs have been used in medical image analysis since the early 1990s for the detection of microcalcifications in digitized mammograms as well as for distinguishing between biopsy‐proven masses and normal tissue on mammograms . More recently, deep learning methods have allowed for the computer‐aided detection of breast lesions in breast MRI, ultrasound, and mammography …”