Looking for Abnormalities in Mammograms With Self- and Weakly Supervised Reconstruction
- 8 January 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 40 (10), 2711-2722
- https://doi.org/10.1109/tmi.2021.3050040
Abstract
Early breast cancer screening through mammography produces every year millions of images worldwide. Despite the volume of the data generated, these images are not systematically associated with standardized labels. Current protocols encourage giving a malignancy probability to each studied breast but do not require the explicit and burdensome annotation of the affected regions. In this work, we address the problem of abnormality detection in the context of such weakly annotated datasets. We combine domain knowledge about the pathology and clinically available image-wise labels to propose a mixed self- and weakly supervised learning framework for abnormalities reconstruction. We also introduce an auxiliary classification task based on the reconstructed regions to improve explainability. We work with high-resolution imaging that enables our network to capture different findings, including masses, micro-calcifications, distortions, and asymmetries, unlike most state-of-the-art works that mainly focus on masses. We use the popular INBreast dataset as well as our private multi-manufacturer dataset for validation and we challenge our method in segmentation, detection, and classification versus multiple state-of-the-art methods. Our results include image-wise AUC up to 0.86, overall region detection true positives rate of 0.93, and the pixel-wise F1 score of 64% on malignant masses.Keywords
Funding Information
- Hera-MI, SAS
- Association Nationale de la Recherche et de la Technologie (2018/0308)
- European Regional Development Fund, Pays de la Loire and Nantes Métropole
This publication has 42 references indexed in Scilit:
- The Importance of Skip Connections in Biomedical Image SegmentationPublished by Springer Science and Business Media LLC ,2016
- U-Net: Convolutional Networks for Biomedical Image SegmentationPublished by Springer Science and Business Media LLC ,2015
- A computational model to generate simulated three‐dimensional breast massesMedical Physics, 2015
- BI-RADS UpdateRadiologic Clinics of North America, 2014
- Computer-Aided Breast Cancer Detection Using Mammograms: A ReviewIEEE Reviews in Biomedical Engineering, 2012
- INbreastAcademic Radiology, 2012
- Calcification in breast lesions: pathologists' perspectiveJournal of Clinical Pathology, 2007
- BIRADS™ classification in mammographyEuropean Journal of Radiology, 2007
- Twenty-Year Follow-up of a Randomized Trial Comparing Total Mastectomy, Lumpectomy, and Lumpectomy plus Irradiation for the Treatment of Invasive Breast CancerNew England Journal of Medicine, 2002
- Comparing the Areas under Two or More Correlated Receiver Operating Characteristic Curves: A Nonparametric ApproachBiometrics, 1988