Hybrid optimization between iterative and network fine-tuning reconstructions for fast quantitative susceptibility mapping

Jinwei Zhang, Hang Zhang, Pascal Spincemaille, Thanh Nguyen, Mert R. Sabuncu, Yi Wang
Proceedings of the Fourth Conference on Medical Imaging with Deep Learning, PMLR 143:870-880, 2021.

Abstract

A Hybrid Optimization Between Iterative and network fine-Tuning (HOBIT) reconstruction method is proposed to solve quantitative susceptibility mapping (QSM) inverse problem in MRI. In HOBIT, a convolutional neural network (CNN) is first trained on healthy subjects’ data with gold standard labels. Domain adaptation to patients’ data with hemorrhagic lesions is then deployed by minimizing fidelity loss on the patient training dataset. During test time, a fidelity loss is imposed on each patient test case, where alternating direction method of multiplier (ADMM) is used to split the time consuming fidelity imposed network update into iterative reconstruction and network update subproblems alternatively in ADMM, and only a subnet of the pre-trained CNN is updated during the process. Compared to the method FINE where such fidelity imposing strategy was initially proposed to solve QSM, HOBIT achieved both performance gain of reconstruction accuracy and vast reduction of computational time.

Cite this Paper

BibTeX
@InProceedings{pmlr-v143-zhang21c, title = {Hybrid optimization between iterative and network fine-tuning reconstructions for fast quantitative susceptibility mapping}, author = {Zhang, Jinwei and Zhang, Hang and Spincemaille, Pascal and Nguyen, Thanh and Sabuncu, Mert R. and Wang, Yi}, booktitle = {Proceedings of the Fourth Conference on Medical Imaging with Deep Learning}, pages = {870--880}, year = {2021}, editor = {Heinrich, Mattias and Dou, Qi and de Bruijne, Marleen and Lellmann, Jan and Schläfer, Alexander and Ernst, Floris}, volume = {143}, series = {Proceedings of Machine Learning Research}, month = {07--09 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v143/zhang21c/zhang21c.pdf}, url = {https://proceedings.mlr.press/v143/zhang21c.html}, abstract = {A Hybrid Optimization Between Iterative and network fine-Tuning (HOBIT) reconstruction method is proposed to solve quantitative susceptibility mapping (QSM) inverse problem in MRI. In HOBIT, a convolutional neural network (CNN) is first trained on healthy subjects’ data with gold standard labels. Domain adaptation to patients’ data with hemorrhagic lesions is then deployed by minimizing fidelity loss on the patient training dataset. During test time, a fidelity loss is imposed on each patient test case, where alternating direction method of multiplier (ADMM) is used to split the time consuming fidelity imposed network update into iterative reconstruction and network update subproblems alternatively in ADMM, and only a subnet of the pre-trained CNN is updated during the process. Compared to the method FINE where such fidelity imposing strategy was initially proposed to solve QSM, HOBIT achieved both performance gain of reconstruction accuracy and vast reduction of computational time.} }
Endnote
%0 Conference Paper %T Hybrid optimization between iterative and network fine-tuning reconstructions for fast quantitative susceptibility mapping %A Jinwei Zhang %A Hang Zhang %A Pascal Spincemaille %A Thanh Nguyen %A Mert R. Sabuncu %A Yi Wang %B Proceedings of the Fourth Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2021 %E Mattias Heinrich %E Qi Dou %E Marleen de Bruijne %E Jan Lellmann %E Alexander Schläfer %E Floris Ernst %F pmlr-v143-zhang21c %I PMLR %P 870--880 %U https://proceedings.mlr.press/v143/zhang21c.html %V 143 %X A Hybrid Optimization Between Iterative and network fine-Tuning (HOBIT) reconstruction method is proposed to solve quantitative susceptibility mapping (QSM) inverse problem in MRI. In HOBIT, a convolutional neural network (CNN) is first trained on healthy subjects’ data with gold standard labels. Domain adaptation to patients’ data with hemorrhagic lesions is then deployed by minimizing fidelity loss on the patient training dataset. During test time, a fidelity loss is imposed on each patient test case, where alternating direction method of multiplier (ADMM) is used to split the time consuming fidelity imposed network update into iterative reconstruction and network update subproblems alternatively in ADMM, and only a subnet of the pre-trained CNN is updated during the process. Compared to the method FINE where such fidelity imposing strategy was initially proposed to solve QSM, HOBIT achieved both performance gain of reconstruction accuracy and vast reduction of computational time.
APA
Zhang, J., Zhang, H., Spincemaille, P., Nguyen, T., Sabuncu, M.R. & Wang, Y.. (2021). Hybrid optimization between iterative and network fine-tuning reconstructions for fast quantitative susceptibility mapping. Proceedings of the Fourth Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 143:870-880 Available from https://proceedings.mlr.press/v143/zhang21c.html.

Related Material