Warmstart Approach for Accelerating Deep Image Prior Reconstruction in Dynamic Tomography

Tobias Knopp, Mirco Grosser
Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, PMLR 172:713-725, 2022.

Abstract

Deep image prior (DIP) has been successfully used in the field of tomography to obtain high-quality images from under-sampled and noisy measurements. The key advantage of DIP compared to conventional deep-learning based image reconstruction techniques is that it requires no training data and thus can be used in a flexible manner without incorporating domain specific knowledge. The downside of DIP is that it shifts the training step to reconstruction time where usually fast algorithms are required to reduced the latency between acquisition and display of the reconstructed image. In this work we tackle this problem for dynamic tomography scenarios in which a large number of temporally resolved images are taken over time. By initializing the DIP network using a previous frame of the time series, it is possible to significantly reduce the overall reconstruction time. To cope with abrupt changes in the captured time-series, we propose to use an adaptive restart method having the ability to switch between warm- and coldstart depending on the amount of inter-frame changes.

Cite this Paper

BibTeX
@InProceedings{pmlr-v172-knopp22a, title = {Warmstart Approach for Accelerating Deep Image Prior Reconstruction in Dynamic Tomography}, author = {Knopp, Tobias and Grosser, Mirco}, booktitle = {Proceedings of The 5th International Conference on Medical Imaging with Deep Learning}, pages = {713--725}, year = {2022}, editor = {Konukoglu, Ender and Menze, Bjoern and Venkataraman, Archana and Baumgartner, Christian and Dou, Qi and Albarqouni, Shadi}, volume = {172}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v172/knopp22a/knopp22a.pdf}, url = {https://proceedings.mlr.press/v172/knopp22a.html}, abstract = {Deep image prior (DIP) has been successfully used in the field of tomography to obtain high-quality images from under-sampled and noisy measurements. The key advantage of DIP compared to conventional deep-learning based image reconstruction techniques is that it requires no training data and thus can be used in a flexible manner without incorporating domain specific knowledge. The downside of DIP is that it shifts the training step to reconstruction time where usually fast algorithms are required to reduced the latency between acquisition and display of the reconstructed image. In this work we tackle this problem for dynamic tomography scenarios in which a large number of temporally resolved images are taken over time. By initializing the DIP network using a previous frame of the time series, it is possible to significantly reduce the overall reconstruction time. To cope with abrupt changes in the captured time-series, we propose to use an adaptive restart method having the ability to switch between warm- and coldstart depending on the amount of inter-frame changes.} }
Endnote
%0 Conference Paper %T Warmstart Approach for Accelerating Deep Image Prior Reconstruction in Dynamic Tomography %A Tobias Knopp %A Mirco Grosser %B Proceedings of The 5th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2022 %E Ender Konukoglu %E Bjoern Menze %E Archana Venkataraman %E Christian Baumgartner %E Qi Dou %E Shadi Albarqouni %F pmlr-v172-knopp22a %I PMLR %P 713--725 %U https://proceedings.mlr.press/v172/knopp22a.html %V 172 %X Deep image prior (DIP) has been successfully used in the field of tomography to obtain high-quality images from under-sampled and noisy measurements. The key advantage of DIP compared to conventional deep-learning based image reconstruction techniques is that it requires no training data and thus can be used in a flexible manner without incorporating domain specific knowledge. The downside of DIP is that it shifts the training step to reconstruction time where usually fast algorithms are required to reduced the latency between acquisition and display of the reconstructed image. In this work we tackle this problem for dynamic tomography scenarios in which a large number of temporally resolved images are taken over time. By initializing the DIP network using a previous frame of the time series, it is possible to significantly reduce the overall reconstruction time. To cope with abrupt changes in the captured time-series, we propose to use an adaptive restart method having the ability to switch between warm- and coldstart depending on the amount of inter-frame changes.
APA
Knopp, T. & Grosser, M.. (2022). Warmstart Approach for Accelerating Deep Image Prior Reconstruction in Dynamic Tomography. Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 172:713-725 Available from https://proceedings.mlr.press/v172/knopp22a.html.

Related Material