Transfer Learning for Image Segmentation by Combining Image Weighting and Kernel Learning

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Standard

Transfer Learning for Image Segmentation by Combining Image Weighting and Kernel Learning. / van Opbroek, Annegreet; Achterberg, Hakim C.; Vernooij, Meike W.; de Bruijne, Marleen.

I: IEEE Transactions on Medical Imaging, Bind 38, Nr. 1, 8419778, 2019, s. 213-224.

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Harvard

van Opbroek, A, Achterberg, HC, Vernooij, MW & de Bruijne, M 2019, 'Transfer Learning for Image Segmentation by Combining Image Weighting and Kernel Learning', IEEE Transactions on Medical Imaging, bind 38, nr. 1, 8419778, s. 213-224. https://doi.org/10.1109/TMI.2018.2859478

APA

van Opbroek, A., Achterberg, H. C., Vernooij, M. W., & de Bruijne, M. (2019). Transfer Learning for Image Segmentation by Combining Image Weighting and Kernel Learning. IEEE Transactions on Medical Imaging, 38(1), 213-224. [8419778]. https://doi.org/10.1109/TMI.2018.2859478

Vancouver

van Opbroek A, Achterberg HC, Vernooij MW, de Bruijne M. Transfer Learning for Image Segmentation by Combining Image Weighting and Kernel Learning. IEEE Transactions on Medical Imaging. 2019;38(1):213-224. 8419778. https://doi.org/10.1109/TMI.2018.2859478

Author

van Opbroek, Annegreet ; Achterberg, Hakim C. ; Vernooij, Meike W. ; de Bruijne, Marleen. / Transfer Learning for Image Segmentation by Combining Image Weighting and Kernel Learning. I: IEEE Transactions on Medical Imaging. 2019 ; Bind 38, Nr. 1. s. 213-224.

Bibtex

@article{5dff0c0298bd4eb2b4570eda4ee6cb7e,
title = "Transfer Learning for Image Segmentation by Combining Image Weighting and Kernel Learning",
abstract = "Many medical image segmentation methods are based on supervised classification of voxels. Such methods generally perform well when provided with a training set that is representative of the test images to segment. However, problems may arise when training and test data follow different distributions, for example due to differences in scanners, scanning protocols, or patient groups. Under such conditions, weighting training images according to distribution similarity has been shown to greatly improve performance. However, this assumes that part of the training data is representative of the test data; it does not make unrepresentative data more similar. We therefore investigate kernel learning as a way to reduce differences between training and test data and explore the added value of kernel learning for image weighting. We also propose a new image weighting method that minimizes maximum mean discrepancy (MMD) between training and test data, which enables the joint optimization of image weights and kernel. Experiments on brain tissue, white matter lesion, and hippocampus segmentation show that both kernel learning and image weighting, when used separately, greatly improve performance on heterogeneous data. Here, MMD weighting obtains similar performance to previously proposed image weighting methods. Combining image weighting and kernel learning, optimized either individually or jointly, can give a small additional improvement in performance.",
author = "{van Opbroek}, Annegreet and Achterberg, {Hakim C.} and Vernooij, {Meike W.} and {de Bruijne}, Marleen",
year = "2019",
doi = "10.1109/TMI.2018.2859478",
language = "English",
volume = "38",
pages = "213--224",
journal = "I E E E Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers",
number = "1",

}

RIS

TY - JOUR

T1 - Transfer Learning for Image Segmentation by Combining Image Weighting and Kernel Learning

AU - van Opbroek, Annegreet

AU - Achterberg, Hakim C.

AU - Vernooij, Meike W.

AU - de Bruijne, Marleen

PY - 2019

Y1 - 2019

N2 - Many medical image segmentation methods are based on supervised classification of voxels. Such methods generally perform well when provided with a training set that is representative of the test images to segment. However, problems may arise when training and test data follow different distributions, for example due to differences in scanners, scanning protocols, or patient groups. Under such conditions, weighting training images according to distribution similarity has been shown to greatly improve performance. However, this assumes that part of the training data is representative of the test data; it does not make unrepresentative data more similar. We therefore investigate kernel learning as a way to reduce differences between training and test data and explore the added value of kernel learning for image weighting. We also propose a new image weighting method that minimizes maximum mean discrepancy (MMD) between training and test data, which enables the joint optimization of image weights and kernel. Experiments on brain tissue, white matter lesion, and hippocampus segmentation show that both kernel learning and image weighting, when used separately, greatly improve performance on heterogeneous data. Here, MMD weighting obtains similar performance to previously proposed image weighting methods. Combining image weighting and kernel learning, optimized either individually or jointly, can give a small additional improvement in performance.

AB - Many medical image segmentation methods are based on supervised classification of voxels. Such methods generally perform well when provided with a training set that is representative of the test images to segment. However, problems may arise when training and test data follow different distributions, for example due to differences in scanners, scanning protocols, or patient groups. Under such conditions, weighting training images according to distribution similarity has been shown to greatly improve performance. However, this assumes that part of the training data is representative of the test data; it does not make unrepresentative data more similar. We therefore investigate kernel learning as a way to reduce differences between training and test data and explore the added value of kernel learning for image weighting. We also propose a new image weighting method that minimizes maximum mean discrepancy (MMD) between training and test data, which enables the joint optimization of image weights and kernel. Experiments on brain tissue, white matter lesion, and hippocampus segmentation show that both kernel learning and image weighting, when used separately, greatly improve performance on heterogeneous data. Here, MMD weighting obtains similar performance to previously proposed image weighting methods. Combining image weighting and kernel learning, optimized either individually or jointly, can give a small additional improvement in performance.

UR - http://www.scopus.com/inward/record.url?scp=85050597954&partnerID=8YFLogxK

U2 - 10.1109/TMI.2018.2859478

DO - 10.1109/TMI.2018.2859478

M3 - Journal article

C2 - 30047874

AN - SCOPUS:85050597954

VL - 38

SP - 213

EP - 224

JO - I E E E Transactions on Medical Imaging

JF - I E E E Transactions on Medical Imaging

SN - 0278-0062

IS - 1

M1 - 8419778

ER -

ID: 200825479