Learning a Loss Function for Segmentation: A Feasibility Study
| Autor: | Tomasz Morgas, Bianca Lassen-Schmidt, Jan Hendrik Moltz, Jan Schreier, Jan Klein, Angelo Genghi, Annika Hänsch, Benjamin Haas |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Computer science
business.industry 020207 software engineering Dice Pattern recognition 02 engineering and technology Function (mathematics) Image segmentation 030218 nuclear medicine & medical imaging 03 medical and health sciences 0302 clinical medicine Sørensen–Dice coefficient 0202 electrical engineering electronic engineering information engineering Segmentation Fraction (mathematics) Artificial intelligence Differentiable function business |
| Zdroj: | ISBI |
| DOI: | 10.1109/isbi45749.2020.9098557 |
| Popis: | When training neural networks for segmentation, the Dice loss is typically used. Alternative loss functions could help the networks achieve results with higher user acceptance and lower correction effort, but they cannot be used directly if they are not differentiable. As a solution, we propose to train a regression network to approximate the loss function and combine it with a U-Net to compute the loss during segmentation training. As an example, we introduce the contour Dice coefficient (CDC) that estimates the fraction of contour length that needs correction. Applied to CT bladder segmentation, we show that a weighted combination of Dice and CDC loss improves segmentations compared to using only Dice loss, with regard to both CDC and other metrics. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|