Name
Title | ||
|---|---|---|
A deep learning framework to classify breast density with noisy labels regularization |
Abstract | ||
|---|---|---|
•A preprocess that allows a correct breast segmentation in mammographies with noisy background.•A preprocess that adjust the intensities to eliminate problems such as unusual brightness.•An intuitive preprocess protocol that normalizes the gray level variability caused by different acquisition devices or different capture processes.•The implementation of a convolution-based architecture capable of modeling multiple radiologists’ opinions, thus reducing the existing variability caused by the noisy labels.•The results suggest this system behaves like a radiologist in the task of classifying mammograms according to their breast density. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1016/j.cmpb.2022.106885 | Computer Methods and Programs in Biomedicine |
Keywords | DocType | Volume |
Breast density,Noisy labels,Deep learning,Dense tissue classification,Mammography | Journal | 221 |
ISSN | Citations | PageRank |
0169-2607 | 0 | 0.34 |
References | Authors | |
0 | 9 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Hector Lopez-Almazan | 1 | 0 | 0.34 |
| Francisco Javier Pérez-Benito | 2 | 0 | 0.34 |
| Andrés Larroza | 3 | 0 | 0.34 |
| Juan-Carlos Perez-Cortes | 4 | 0 | 0.34 |
| Marina Pollan | 5 | 0 | 0.34 |
| Beatriz Perez-Gomez | 6 | 0 | 0.34 |
| Dolores Salas Trejo | 7 | 0 | 0.34 |
| María Casals | 8 | 0 | 0.34 |
| Rafael Llobet | 9 | 0 | 0.34 |