Breast Dense Tissue Segmentation with Noisy Labels: A Hybrid Threshold-Based and Mask-Based Approach
Por:
Larroza A, Pérez-Benito FJ, Perez-Cortes JC, Román M, Pollán M, Pérez-Gómez B, Salas-Trejo D, Casals M and Llobet R
Publicada:
1 ago 2022
Ahead of Print:
28 jul 2022
Categoría:
Clinical biochemistry
Resumen:
Breast density assessed from digital mammograms is a known biomarker related to a higher risk of developing breast cancer. Supervised learning algorithms have been implemented to determine this. However, the performance of these algorithms depends on the quality of the ground-truth information, which expert readers usually provide. These expert labels are noisy approximations to the ground truth, as there is both intra- and inter-observer variability among them. Thus, it is crucial to provide a reliable method to measure breast density from mammograms. This paper presents a fully automated method based on deep learning to estimate breast density, including breast detection, pectoral muscle exclusion, and dense tissue segmentation. We propose a novel confusion matrix (CM)-YNet model for the segmentation step. This architecture includes networks to model each radiologist's noisy label and gives the estimated ground-truth segmentation as well as two parameters that allow interaction with a threshold-based labeling tool. A multi-center study involving 1785 women whose "for presentation" mammograms were obtained from 11 different medical facilities was performed. A total of 2496 mammograms were used as the training corpus, and 844 formed the testing corpus. Additionally, we included a totally independent dataset from a different center, composed of 381 women with one image per patient. Each mammogram was labeled independently by two expert radiologists using a threshold-based tool. The implemented CM-Ynet model achieved the highest DICE score averaged over both test datasets (0.82 +/- 0.14) when compared to the closest dense-tissue segmentation assessment from both radiologists. The level of concordance between the two radiologists showed a DICE score of 0.76 +/- 0.17. An automatic breast density estimator based on deep learning exhibited higher performance when compared with two experienced radiologists. This suggests that modeling each radiologist's label allows for better estimation of the unknown ground-truth segmentation. The advantage of the proposed model is that it also provides the threshold parameters that enable user interaction with a threshold-based tool.
Filiaciones:
Larroza A:
Instituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera, 46022 València, Spain
Pérez-Benito FJ:
Instituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera, 46022 València, Spain
Perez-Cortes JC:
Instituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera, 46022 València, Spain
Román M:
Department of Epidemiology and Evaluation, IMIM (Hospital del Mar Medical Research Institute), Passeig Marítim 25-29, 08003 Barcelona, Spain
Pollán M:
National Center for Epidemiology, Carlos III Institute of Health, Monforte de lemos, 5, 28029 Madrid, Spain
Consortium for Biomedical Research in Epidemiology and Public Health (CIBER en Epidemiología y Salud Pública-CIBERESP), Carlos III Institute of Health, Monforte de Lemos, 5, 28029 Madrid, Spain
Pérez-Gómez B:
National Center for Epidemiology, Carlos III Institute of Health, Monforte de lemos, 5, 28029 Madrid, Spain
Consortium for Biomedical Research in Epidemiology and Public Health (CIBER en Epidemiología y Salud Pública-CIBERESP), Carlos III Institute of Health, Monforte de Lemos, 5, 28029 Madrid, Spain
:
Valencian Breast Cancer Screening Program, General Directorate of Public Health, 46022 València, Spain
Centro Superior de Investigación en Salud Pública, CSISP, FISABIO, 46020 València, Spain
:
Valencian Breast Cancer Screening Program, General Directorate of Public Health, 46022 València, Spain
Centro Superior de Investigación en Salud Pública, CSISP, FISABIO, 46020 València, Spain
Llobet R:
Instituto Tecnológico de la Informática, Universitat Politècnica de València, Camino de Vera, 46022 València, Spain
gold, Green Published
|