Share Email Print
cover

Proceedings Paper • new

Using a convolutional neural network to predict readers' estimates of mammographic density for breast cancer risk assessment
Author(s): Georgia V. Ionescu; Martin Fergie; Michael Berks; Elaine F. Harkness; Johan Hulleman; Adam R. Brentnall; Jack Cuzick; D. Gareth Evans; Susan M. Astley
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Background: Mammographic density is an important risk factor for breast cancer. Recent research demonstrated that percentage density assessed visually using Visual Analogue Scales (VAS) showed stronger risk prediction than existing automated density measures, suggesting readers may recognise relevant image features not yet captured by automated methods.

Method: We have built convolutional neural networks (CNN) to predict VAS scores from full-field digital mammograms. The CNNs are trained using whole-image mammograms, each labelled with the average VAS score of two independent readers. They learn a mapping between mammographic appearance and VAS score so that at test time, they can predict VAS score for an unseen image. Networks were trained using 67520 mammographic images from 16968 women, and tested on a large dataset of 73128 images and case-control sets of contralateral mammograms of screen detected cancers and prior images of women with cancers detected subsequently, matched to controls on age, menopausal status, parity, HRT and BMI.

Results: Pearson's correlation coefficient between readers' and predicted VAS in the large dataset was 0.79 per mammogram and 0.83 per woman (averaging over all views). In the case-control sets, odds ratios of cancer in the highest vs lowest quintile of percentage density were 3.07 (95%CI: 1.97 - 4.77) for the screen detected cancers and 3.52 (2.22 - 5.58) for the priors, with matched concordance indices of 0.59 (0.55 - 0.64) and 0.61 (0.58 - 0.65) respectively.

Conclusion: Our fully automated method demonstrated encouraging results which compare well with existing methods, including VAS.

Paper Details

Date Published: 6 July 2018
PDF: 10 pages
Proc. SPIE 10718, 14th International Workshop on Breast Imaging (IWBI 2018), 107180D (6 July 2018); doi: 10.1117/12.2318464
Show Author Affiliations
Georgia V. Ionescu, The Univ. of Manchester (United Kingdom)
Martin Fergie, The Univ. of Manchester (United Kingdom)
Michael Berks, The Univ. of Manchester (United Kingdom)
Elaine F. Harkness, The Univ. of Manchester (United Kingdom)
Manchester Univ. NHS Foundation Trust (United Kingdom)
Johan Hulleman, The Univ. of Manchester (United Kingdom)
Adam R. Brentnall, Queen Mary Univ. of London (United Kingdom)
Jack Cuzick, Queen Mary Univ. of London (United Kingdom)
D. Gareth Evans, Manchester Univ. NHS Foundation Trust (United Kingdom)
The Christie NHS Foundation Trust (United Kingdom)
The Univ. of Manchester (United Kingdom)
Susan M. Astley, The Univ. of Manchester (United Kingdom)
Manchester Univ. NHS Foundation Trust (United Kingdom)


Published in SPIE Proceedings Vol. 10718:
14th International Workshop on Breast Imaging (IWBI 2018)
Elizabeth A. Krupinski, Editor(s)

© SPIE. Terms of Use
Back to Top