Instance-Specific Augmentation of Brain MRIs with Variational Autoencoders

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Standard

Instance-Specific Augmentation of Brain MRIs with Variational Autoencoders. / Middleton, Jon Anthony; Bauer, Marko; Johansen, Jacob; Nielsen, Mads; Sommer, Stefan Horst; Pai, Akshay Sadananda Uppinakudru.

Medical Applications with Disentanglements . 2023.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Harvard

Middleton, JA, Bauer, M, Johansen, J, Nielsen, M, Sommer, SH & Pai, ASU 2023, Instance-Specific Augmentation of Brain MRIs with Variational Autoencoders. in Medical Applications with Disentanglements . https://doi.org/10.1007/978-3-031-25046-0_5

APA

Middleton, J. A., Bauer, M., Johansen, J., Nielsen, M., Sommer, S. H., & Pai, A. S. U. (2023). Instance-Specific Augmentation of Brain MRIs with Variational Autoencoders. In Medical Applications with Disentanglements https://doi.org/10.1007/978-3-031-25046-0_5

Vancouver

Middleton JA, Bauer M, Johansen J, Nielsen M, Sommer SH, Pai ASU. Instance-Specific Augmentation of Brain MRIs with Variational Autoencoders. In Medical Applications with Disentanglements . 2023 https://doi.org/10.1007/978-3-031-25046-0_5

Author

Middleton, Jon Anthony ; Bauer, Marko ; Johansen, Jacob ; Nielsen, Mads ; Sommer, Stefan Horst ; Pai, Akshay Sadananda Uppinakudru. / Instance-Specific Augmentation of Brain MRIs with Variational Autoencoders. Medical Applications with Disentanglements . 2023.

Bibtex

@inproceedings{58726862d1514d97b32f2e5ac69b568f,
title = "Instance-Specific Augmentation of Brain MRIs with Variational Autoencoders",
abstract = "Spatial data augmentation is a standard technique for regularizing deep segmentation networks that are tasked with localizing medical abnormalities. However, a typical spatial augmentation scheme is built upon ad hoc selections of spatial transformation parameters which are not determined by the data set and therefore may not capture spatial variations in the data. For segmentation networks trained in the low-data regime, these ad hoc transformation techniques often fail to encourage better generalization. To address this problem, we propose a variational autoencoder framework for spatial data augmentation. We show how this framework provides a natural, data-driven approach to probabilistic, instance-specific spatial augmentation. Further, we observe that U-Nets trained on data augmented using this framework compare favorably with U-Nets trained using standard spatial augmentation methods.",
author = "Middleton, {Jon Anthony} and Marko Bauer and Jacob Johansen and Mads Nielsen and Sommer, {Stefan Horst} and Pai, {Akshay Sadananda Uppinakudru}",
year = "2023",
doi = "https://doi.org/10.1007/978-3-031-25046-0_5",
language = "English",
booktitle = "Medical Applications with Disentanglements",

}

RIS

TY - GEN

T1 - Instance-Specific Augmentation of Brain MRIs with Variational Autoencoders

AU - Middleton, Jon Anthony

AU - Bauer, Marko

AU - Johansen, Jacob

AU - Nielsen, Mads

AU - Sommer, Stefan Horst

AU - Pai, Akshay Sadananda Uppinakudru

PY - 2023

Y1 - 2023

N2 - Spatial data augmentation is a standard technique for regularizing deep segmentation networks that are tasked with localizing medical abnormalities. However, a typical spatial augmentation scheme is built upon ad hoc selections of spatial transformation parameters which are not determined by the data set and therefore may not capture spatial variations in the data. For segmentation networks trained in the low-data regime, these ad hoc transformation techniques often fail to encourage better generalization. To address this problem, we propose a variational autoencoder framework for spatial data augmentation. We show how this framework provides a natural, data-driven approach to probabilistic, instance-specific spatial augmentation. Further, we observe that U-Nets trained on data augmented using this framework compare favorably with U-Nets trained using standard spatial augmentation methods.

AB - Spatial data augmentation is a standard technique for regularizing deep segmentation networks that are tasked with localizing medical abnormalities. However, a typical spatial augmentation scheme is built upon ad hoc selections of spatial transformation parameters which are not determined by the data set and therefore may not capture spatial variations in the data. For segmentation networks trained in the low-data regime, these ad hoc transformation techniques often fail to encourage better generalization. To address this problem, we propose a variational autoencoder framework for spatial data augmentation. We show how this framework provides a natural, data-driven approach to probabilistic, instance-specific spatial augmentation. Further, we observe that U-Nets trained on data augmented using this framework compare favorably with U-Nets trained using standard spatial augmentation methods.

U2 - https://doi.org/10.1007/978-3-031-25046-0_5

DO - https://doi.org/10.1007/978-3-031-25046-0_5

M3 - Article in proceedings

BT - Medical Applications with Disentanglements

ER -

ID: 334464739