TY - GEN
T1 - Effect of statistical mismatch between training and test images for CNN-based deformable registration
AU - Ketcha, M. D.
AU - De Silva, T.
AU - Han, R.
AU - Uneri, A.
AU - Vogt, S.
AU - Kleinszig, G.
AU - Siewerdsen, J. H.
N1 - Funding Information:
Acknowledgment: Research supported by NIH Grant No. R01-EB-017226 and collaboration with Siemens XP.
Publisher Copyright:
© COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.
PY - 2019
Y1 - 2019
N2 - Recently, convolutional neural networks (CNNs) have been proposed as a method for deformable image registration, offering a variety of potential advantages compared to physical model-based methods, including faster runtime and ability to learn complicated functions without explicit models. A persistent question for CNNs is the uncertainty in their behavior when the image statistics (e.g., noise and resolution) of the test data deviate from those of the training data. In this work we investigated the influence of statistical properties of image noise (in CT, for example, related to radiation dose) and deformation magnitude, trained registration networks over a range of dose and deformation levels, and evaluated registration performance (target registration error, TRE) as the statistics of the test data deviated from that of the training data. Generally, registration performance was optimal when the statistics of the test data matched that of the training data, except in cases of very low-dose data, where networks trained on a combination of high- and low-dose images achieved best TRE. Furthermore, TRE was found to be limited by the highest dose training data, with no improvement in TRE for test images of higher dose than that in the training data. Understanding and quantifying the relationship between statistical aspects of the training and test data - and the failure modes caused by statistical mismatch - is an important step in the development of CNN-based registration methods. This work provided new insight on the optima and tradeoffs with respect to image noise (dose) and deformation magnitude, providing important guidance in building training sets that are bestsuited to particular imaging conditions and applications.
AB - Recently, convolutional neural networks (CNNs) have been proposed as a method for deformable image registration, offering a variety of potential advantages compared to physical model-based methods, including faster runtime and ability to learn complicated functions without explicit models. A persistent question for CNNs is the uncertainty in their behavior when the image statistics (e.g., noise and resolution) of the test data deviate from those of the training data. In this work we investigated the influence of statistical properties of image noise (in CT, for example, related to radiation dose) and deformation magnitude, trained registration networks over a range of dose and deformation levels, and evaluated registration performance (target registration error, TRE) as the statistics of the test data deviated from that of the training data. Generally, registration performance was optimal when the statistics of the test data matched that of the training data, except in cases of very low-dose data, where networks trained on a combination of high- and low-dose images achieved best TRE. Furthermore, TRE was found to be limited by the highest dose training data, with no improvement in TRE for test images of higher dose than that in the training data. Understanding and quantifying the relationship between statistical aspects of the training and test data - and the failure modes caused by statistical mismatch - is an important step in the development of CNN-based registration methods. This work provided new insight on the optima and tradeoffs with respect to image noise (dose) and deformation magnitude, providing important guidance in building training sets that are bestsuited to particular imaging conditions and applications.
KW - Convolutional Neural Networks
KW - Deformable Image Registration
KW - Image Statistics
UR - http://www.scopus.com/inward/record.url?scp=85068335070&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85068335070&partnerID=8YFLogxK
U2 - 10.1117/12.2512824
DO - 10.1117/12.2512824
M3 - Conference contribution
AN - SCOPUS:85068335070
T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE
BT - Medical Imaging 2019
A2 - Angelini, Elsa D.
A2 - Angelini, Elsa D.
A2 - Angelini, Elsa D.
A2 - Landman, Bennett A.
PB - SPIE
T2 - Medical Imaging 2019: Image Processing
Y2 - 19 February 2019 through 21 February 2019
ER -