Robust abdominal organ segmentation using regional convolutional neural networks
(2017) 20th Scandinavian Conference on Image Analysis, SCIA 2017 In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 10270 LNCS. p.41-52- 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. The convolutional neural network consists of several full 3D convolutional layers and takes both low and high resolution image data as input, which is designed to ensure both local and global consistency. 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... (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. The convolutional neural network consists of several full 3D convolutional layers and takes both low and high resolution image data as input, which is designed to ensure both local and global consistency. 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. It achieved the best results for 3 out of the 13 organs with a total mean Dice coefficient of 0.757 for all organs. Top scores were achieved for the gallbladder, the aorta and the right adrenal gland.
(Less)
- author
- Larsson, Måns ; Zhang, Yuhang and Kahl, Fredrik LU
- organization
- publishing date
- 2017
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Convolutional neural networks, Medical image analysis, Segmentation
- host publication
- Image Analysis - 20th Scandinavian Conference, SCIA 2017, Proceedings
- series title
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
- volume
- 10270 LNCS
- pages
- 12 pages
- publisher
- Springer
- conference name
- 20th Scandinavian Conference on Image Analysis, SCIA 2017
- conference location
- Tromso, Norway
- conference dates
- 2017-06-12 - 2017-06-14
- external identifiers
-
- scopus:85020380711
- ISSN
- 16113349
- 03029743
- ISBN
- 9783319591285
- DOI
- 10.1007/978-3-319-59129-2_4
- language
- English
- LU publication?
- yes
- id
- aa51f526-4e38-459d-91b7-77ebdfcd95bb
- date added to LUP
- 2017-06-30 09:03:19
- date last changed
- 2024-10-14 08:51:18
@inproceedings{aa51f526-4e38-459d-91b7-77ebdfcd95bb, 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. The convolutional neural network consists of several full 3D convolutional layers and takes both low and high resolution image data as input, which is designed to ensure both local and global consistency. 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. It achieved the best results for 3 out of the 13 organs with a total mean Dice coefficient of 0.757 for all organs. Top scores were achieved for the gallbladder, the aorta and the right adrenal gland.</p>}}, author = {{Larsson, Måns and Zhang, Yuhang and Kahl, Fredrik}}, booktitle = {{Image Analysis - 20th Scandinavian Conference, SCIA 2017, Proceedings}}, isbn = {{9783319591285}}, issn = {{16113349}}, keywords = {{Convolutional neural networks; Medical image analysis; Segmentation}}, language = {{eng}}, pages = {{41--52}}, publisher = {{Springer}}, series = {{Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)}}, title = {{Robust abdominal organ segmentation using regional convolutional neural networks}}, url = {{http://dx.doi.org/10.1007/978-3-319-59129-2_4}}, doi = {{10.1007/978-3-319-59129-2_4}}, volume = {{10270 LNCS}}, year = {{2017}}, }