Digital Mammography

By Andrew D. A. Maidment. 2019

. Conventional risk models for the development of breast cancer use inputs such as age, weight, hormonal factors and family history to compute individual breast cancer risk. These are employed in the management of women at high risk. The addition of breast density as an input has been shown to im-prove the accuracy of such models. An improved risk model could facilitate risk-based population screening. However, in order to use breast density in risk models there is a need to employ objective methods for measuring the density. A feasibility study has been carried out to assess the practicality of using a stepwedge-based technique for measuring breast density from mammograms in the UK National Health Service Breast Screening Programme and to determine whether additional information, relevant to risk, can be collected by questionnaire. Preliminary results suggest that it is practical to use such a technique in the screening environment. In a sample of 100 women, the mean density was 27% (range 2 - 81%). A negative trend in breast density was observed with Body Mass Index.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Multi-task transfer learning deep convolutional neural network: Application to computer-aided diagnosis of breast cancer on mammograms (2017, methodology): “The heterogeneous data from DDSM and DM are added into each N -fold SFM data during transfer learning.”