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.

No abstract

The main objectives were to describe the measures taken by women to detect breast disease prior to invitation to participate in a screening programme for breast cancer, and to identify factors related to non-participation in this programme. A cross-sectional study was designed at the Breast Cancer Early Detection Program of Sabadell-Cerdanyola (BCEDP), using data collected in interviews conducted face to face or over the telephone with 13 760 women participating in the programme and 280 non-participants. A total of 74.2% of the participants versus 70.4% of the non-participants reported having taken measures to detect breast disease, and 71.7% of the participants had undergone mammography versus 69.6% of the non-participants. Of the 10 057 women who had had mammograms, 58% had done so less than 2 years previously. Factors found to be associated to non-participation in the multivariate analysis were: higher level of education, higher occupational skills or working at home, self- or gynaecological examination of breasts, and having received hormone replacement therapy. Higher age group was the only factor that increased the probability of not having undergone mammography previously. Despite the high prevalence of prior measures to detect breast cancer and the similar prevalence between participating and non-participating women, this behaviour is much less prevalent in the group of women 60 years of age or older.

Chemical shift GRE MR imaging can be used to detect fatty metamorphosis in HCC.

No abstract