Robust abdominal organ segmentation using regional convolutional neural networks
(2018) In Applied Soft Computing 70. p.465-471- Abstract
A fully automatic system for abdominal organ segmentation is presented. As a first step, an organ localization is obtained via a robust and efficient feature registration method where the center of the organ is estimated together with a region of interest surrounding the center. Then, a convolutional neural network performing voxelwise classification is applied. Two convolutional neural networks of different architecture are compared. The first one has a structure similar to networks used for classification and is applied using a sliding window approach. The second one has a structure allowing it to be applied in a fully convolutional manner reducing computation time. Despite limited training data, our experimental results are on par... (More)
A fully automatic system for abdominal organ segmentation is presented. As a first step, an organ localization is obtained via a robust and efficient feature registration method where the center of the organ is estimated together with a region of interest surrounding the center. Then, a convolutional neural network performing voxelwise classification is applied. Two convolutional neural networks of different architecture are compared. The first one has a structure similar to networks used for classification and is applied using a sliding window approach. The second one has a structure allowing it to be applied in a fully convolutional manner reducing computation time. Despite limited training data, our experimental results are on par with state-of-the-art approaches that have been developed over many years. More specifically the method is applied to the MICCAI2015 challenge “Multi-Atlas Labeling Beyond the Cranial Vault” in the free competition for organ segmentation in the abdomen. The method performed well for both types of convolutional neural networks. For the fully convolutional network a mean Dice coefficient of 0.767 was achieved, for the network applied with sliding window this number was 0.757.
(Less)
- author
- Larsson, Måns ; Zhang, Yuhang and Kahl, Fredrik LU
- organization
- publishing date
- 2018-09-01
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Convolutional neural networks, Medical image analysis, Segmentation
- in
- Applied Soft Computing
- volume
- 70
- pages
- 7 pages
- publisher
- Elsevier
- external identifiers
-
- scopus:85048004710
- ISSN
- 1568-4946
- DOI
- 10.1016/j.asoc.2018.05.038
- language
- English
- LU publication?
- yes
- id
- 9269a7af-d846-4447-ae6f-6e6552232933
- date added to LUP
- 2018-06-11 10:16:16
- date last changed
- 2022-04-25 07:46:34
@article{9269a7af-d846-4447-ae6f-6e6552232933,
abstract = {{<p>A fully automatic system for abdominal organ segmentation is presented. As a first step, an organ localization is obtained via a robust and efficient feature registration method where the center of the organ is estimated together with a region of interest surrounding the center. Then, a convolutional neural network performing voxelwise classification is applied. Two convolutional neural networks of different architecture are compared. The first one has a structure similar to networks used for classification and is applied using a sliding window approach. The second one has a structure allowing it to be applied in a fully convolutional manner reducing computation time. Despite limited training data, our experimental results are on par with state-of-the-art approaches that have been developed over many years. More specifically the method is applied to the MICCAI2015 challenge “Multi-Atlas Labeling Beyond the Cranial Vault” in the free competition for organ segmentation in the abdomen. The method performed well for both types of convolutional neural networks. For the fully convolutional network a mean Dice coefficient of 0.767 was achieved, for the network applied with sliding window this number was 0.757.</p>}},
author = {{Larsson, Måns and Zhang, Yuhang and Kahl, Fredrik}},
issn = {{1568-4946}},
keywords = {{Convolutional neural networks; Medical image analysis; Segmentation}},
language = {{eng}},
month = {{09}},
pages = {{465--471}},
publisher = {{Elsevier}},
series = {{Applied Soft Computing}},
title = {{Robust abdominal organ segmentation using regional convolutional neural networks}},
url = {{http://dx.doi.org/10.1016/j.asoc.2018.05.038}},
doi = {{10.1016/j.asoc.2018.05.038}},
volume = {{70}},
year = {{2018}},
}