Unsupervised deep learning applied to breast density segmentation and mammographic risk scoring
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Unsupervised deep learning applied to breast density segmentation and mammographic risk scoring. / Kallenberg, Michiel Gijsbertus J.; Petersen, Peter Kersten; Nielsen, Mads; Ng, Andrew Y.; Diao, Pengfei; Igel, Christian; Vachon, Celine M.; Holland, Katharina; Winkel, Rikke Rass; Karssemeijer, Nico; Lillholm, Martin.
In: IEEE Transactions on Medical Imaging, Vol. 35, No. 5, 2016, p. 1322-1331.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Unsupervised deep learning applied to breast density segmentation and mammographic risk scoring
AU - Kallenberg, Michiel Gijsbertus J.
AU - Petersen, Peter Kersten
AU - Nielsen, Mads
AU - Ng, Andrew Y.
AU - Diao, Pengfei
AU - Igel, Christian
AU - Vachon, Celine M.
AU - Holland, Katharina
AU - Winkel, Rikke Rass
AU - Karssemeijer, Nico
AU - Lillholm, Martin
PY - 2016
Y1 - 2016
N2 - Mammographic risk scoring has commonly been automated by extracting a set of handcrafted features from mammograms, and relating the responses directly or indirectly to breast cancer risk. We present a method that learns a feature hierarchy from unlabeled data. When the learned features are used as the input to a simple classifier, two different tasks can be addressed: i) breast density segmentation, and ii) scoring of mammographic texture. The proposed model learns features at multiple scales. To control the models capacity a novel sparsity regularizer is introduced that incorporates both lifetime and population sparsity. We evaluated our method on three different clinical datasets. Our state-of-the-art results show that the learned breast density scores have a very strong positive relationship with manual ones, and that the learned texture scores are predictive of breast cancer. The model is easy to apply and generalizes to many other segmentation and scoring problems.
AB - Mammographic risk scoring has commonly been automated by extracting a set of handcrafted features from mammograms, and relating the responses directly or indirectly to breast cancer risk. We present a method that learns a feature hierarchy from unlabeled data. When the learned features are used as the input to a simple classifier, two different tasks can be addressed: i) breast density segmentation, and ii) scoring of mammographic texture. The proposed model learns features at multiple scales. To control the models capacity a novel sparsity regularizer is introduced that incorporates both lifetime and population sparsity. We evaluated our method on three different clinical datasets. Our state-of-the-art results show that the learned breast density scores have a very strong positive relationship with manual ones, and that the learned texture scores are predictive of breast cancer. The model is easy to apply and generalizes to many other segmentation and scoring problems.
U2 - 10.1109/TMI.2016.2532122
DO - 10.1109/TMI.2016.2532122
M3 - Journal article
C2 - 26915120
VL - 35
SP - 1322
EP - 1331
JO - I E E E Transactions on Medical Imaging
JF - I E E E Transactions on Medical Imaging
SN - 0278-0062
IS - 5
ER -
ID: 160401322