Deep learning-based parameter mapping for joint relaxation and diffusion tensor MR Fingerprinting

Carolin M. Pirk, Pedro A. Gomez, Ilona Lipp, Guido Buonincontri, Miguel Molina-Romero, Anjany Sekuboyina, Diana Waldmannstetter, Jonathan Dannenberg, Sebastian Endt, Alberto Merola, Joseph R. Whittaker, Valentina Tomassini, Michela Tosetti, Derek K. Jones, Bjoern H. Menze, Marion Menzel
Proceedings of the Third Conference on Medical Imaging with Deep Learning, PMLR 121:638-654, 2020.

Abstract

Magnetic Resonance Fingerprinting (MRF) enables the simultaneous quantification of multiple properties of biological tissues. It relies on a pseudo-random acquisition and the matching of acquired signal evolutions to a precomputed dictionary. However, the dictionary is not scalable to higher-parametric spaces, limiting MRF to the simultaneous mapping of only a small number of parameters (proton density, T1 and T2 in general). Inspired by diffusion-weighted SSFP imaging, we present a proof-of-concept of a novel MRF sequence with embedded diffusion-encoding gradients along all three axes to efficiently encode orientational diffusion and T1 and T2 relaxation. We take advantage of a convolutional neural network (CNN) to reconstruct multiple quantitative maps from this single, highly undersampled acquisition. We bypass expensive dictionary matching by learning the implicit physical relationships between the spatiotemporal MRF data and the T1, T2 and diffusion tensor parameters. The predicted parameter maps and the derived scalar diffusion metrics agree well with state-of-the-art reference protocols. Orientational diffusion information is captured as seen from the estimated primary diffusion directions. In addition to this, the joint acquisition and reconstruction framework proves capable of preserving tissue abnormalities in multiple sclerosis lesions.

Cite this Paper

BibTeX
@InProceedings{pmlr-v121-pirk20a, title = {Deep learning-based parameter mapping for joint relaxation and diffusion tensor MR Fingerprinting}, author = {Pirk, Carolin M. and Gomez, Pedro A. and Lipp, Ilona and Buonincontri, Guido and Molina-Romero, Miguel and Sekuboyina, Anjany and Waldmannstetter, Diana and Dannenberg, Jonathan and Endt, Sebastian and Merola, Alberto and Whittaker, Joseph R. and Tomassini, Valentina and Tosetti, Michela and Jones, Derek K. and Menze, Bjoern H. and Menzel, Marion}, booktitle = {Proceedings of the Third Conference on Medical Imaging with Deep Learning}, pages = {638--654}, year = {2020}, editor = {Arbel, Tal and Ben Ayed, Ismail and de Bruijne, Marleen and Descoteaux, Maxime and Lombaert, Herve and Pal, Christopher}, volume = {121}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v121/pirk20a/pirk20a.pdf}, url = {https://proceedings.mlr.press/v121/pirk20a.html}, abstract = {Magnetic Resonance Fingerprinting (MRF) enables the simultaneous quantification of multiple properties of biological tissues. It relies on a pseudo-random acquisition and the matching of acquired signal evolutions to a precomputed dictionary. However, the dictionary is not scalable to higher-parametric spaces, limiting MRF to the simultaneous mapping of only a small number of parameters (proton density, T1 and T2 in general). Inspired by diffusion-weighted SSFP imaging, we present a proof-of-concept of a novel MRF sequence with embedded diffusion-encoding gradients along all three axes to efficiently encode orientational diffusion and T1 and T2 relaxation. We take advantage of a convolutional neural network (CNN) to reconstruct multiple quantitative maps from this single, highly undersampled acquisition. We bypass expensive dictionary matching by learning the implicit physical relationships between the spatiotemporal MRF data and the T1, T2 and diffusion tensor parameters. The predicted parameter maps and the derived scalar diffusion metrics agree well with state-of-the-art reference protocols. Orientational diffusion information is captured as seen from the estimated primary diffusion directions. In addition to this, the joint acquisition and reconstruction framework proves capable of preserving tissue abnormalities in multiple sclerosis lesions.} }
Endnote
%0 Conference Paper %T Deep learning-based parameter mapping for joint relaxation and diffusion tensor MR Fingerprinting %A Carolin M. Pirk %A Pedro A. Gomez %A Ilona Lipp %A Guido Buonincontri %A Miguel Molina-Romero %A Anjany Sekuboyina %A Diana Waldmannstetter %A Jonathan Dannenberg %A Sebastian Endt %A Alberto Merola %A Joseph R. Whittaker %A Valentina Tomassini %A Michela Tosetti %A Derek K. Jones %A Bjoern H. Menze %A Marion Menzel %B Proceedings of the Third Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2020 %E Tal Arbel %E Ismail Ben Ayed %E Marleen de Bruijne %E Maxime Descoteaux %E Herve Lombaert %E Christopher Pal %F pmlr-v121-pirk20a %I PMLR %P 638--654 %U https://proceedings.mlr.press/v121/pirk20a.html %V 121 %X Magnetic Resonance Fingerprinting (MRF) enables the simultaneous quantification of multiple properties of biological tissues. It relies on a pseudo-random acquisition and the matching of acquired signal evolutions to a precomputed dictionary. However, the dictionary is not scalable to higher-parametric spaces, limiting MRF to the simultaneous mapping of only a small number of parameters (proton density, T1 and T2 in general). Inspired by diffusion-weighted SSFP imaging, we present a proof-of-concept of a novel MRF sequence with embedded diffusion-encoding gradients along all three axes to efficiently encode orientational diffusion and T1 and T2 relaxation. We take advantage of a convolutional neural network (CNN) to reconstruct multiple quantitative maps from this single, highly undersampled acquisition. We bypass expensive dictionary matching by learning the implicit physical relationships between the spatiotemporal MRF data and the T1, T2 and diffusion tensor parameters. The predicted parameter maps and the derived scalar diffusion metrics agree well with state-of-the-art reference protocols. Orientational diffusion information is captured as seen from the estimated primary diffusion directions. In addition to this, the joint acquisition and reconstruction framework proves capable of preserving tissue abnormalities in multiple sclerosis lesions.
APA
Pirk, C.M., Gomez, P.A., Lipp, I., Buonincontri, G., Molina-Romero, M., Sekuboyina, A., Waldmannstetter, D., Dannenberg, J., Endt, S., Merola, A., Whittaker, J.R., Tomassini, V., Tosetti, M., Jones, D.K., Menze, B.H. & Menzel, M.. (2020). Deep learning-based parameter mapping for joint relaxation and diffusion tensor MR Fingerprinting. Proceedings of the Third Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 121:638-654 Available from https://proceedings.mlr.press/v121/pirk20a.html.

Related Material