Subject-Specific Convolutional Neural Networks for Accelerated Magnetic Resonance Imaging

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Magnetic Resonance Imaging (MRI) is one of the leading modalities for medical imaging, providing excellent softtissue contrast without exposure to ionizing radiation. Despite continuing advances in MRI, long scan times remain a major limitation in clinical applications. Parallel imaging is a technique for scan time acceleration in MRI, which utilizes the spatial variations in the reception profiles of receiver coil arrays to reconstruct images from undersampled Fourier space, i.e. K-space. One of the most commonly used parallel imaging techniques employs interpolation of missing k-space information by using linear shiftinvariant convolutional kernels. These kernels are trained on a limited amount of autocalibration signal (ACS) for each scan. We propose a novel method for parallel imaging,Robust Artificialneural-networks for k -space Interpolation (RAKI), which uses scan-specific convolutional neural networks (CNNs) to perform improved k-space interpolation. Three-layer CNNs are trained using only scan-specific ACS data, alleviating the need for large training databases. The proposed method was tested in ultra-high resolution brain MRI and quantitative cardiac MRI, acquired with various acceleration rates. Improved noise resilience as compared to existing parallel imaging methods was observed for high acceleration rates or in the presence of low signal-tonoise ratio (SNR). Furthermore, RAKI successfully reconstructed images for quantitative cardiac MRI, even when using the same CNN across images with varying contrasts. These results indicate that RAKI achieves improved noise performance without overfitting to specific image contents, and offers great promise for improved acceleration in a wide range of MRI applications.

Original languageEnglish (US)
Title of host publication2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509060146
DOIs
StatePublished - Oct 10 2018
Event2018 International Joint Conference on Neural Networks, IJCNN 2018 - Rio de Janeiro, Brazil
Duration: Jul 8 2018Jul 13 2018

Publication series

NameProceedings of the International Joint Conference on Neural Networks
Volume2018-July

Other

Other2018 International Joint Conference on Neural Networks, IJCNN 2018
CountryBrazil
CityRio de Janeiro
Period7/8/187/13/18

Fingerprint

Neural networks
Interpolation
Imaging techniques
Ionizing radiation
Medical imaging
Magnetic Resonance Imaging
Brain

Cite this

Akcakaya, M., Moeller, S., Weingartner, S., & Ugurbil, K. (2018). Subject-Specific Convolutional Neural Networks for Accelerated Magnetic Resonance Imaging. In 2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings [8489393] (Proceedings of the International Joint Conference on Neural Networks; Vol. 2018-July). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IJCNN.2018.8489393

Subject-Specific Convolutional Neural Networks for Accelerated Magnetic Resonance Imaging. / Akcakaya, Mehmet; Moeller, Steen; Weingartner, Sebastian; Ugurbil, Kamil.

2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. 8489393 (Proceedings of the International Joint Conference on Neural Networks; Vol. 2018-July).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Akcakaya, M, Moeller, S, Weingartner, S & Ugurbil, K 2018, Subject-Specific Convolutional Neural Networks for Accelerated Magnetic Resonance Imaging. in 2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings., 8489393, Proceedings of the International Joint Conference on Neural Networks, vol. 2018-July, Institute of Electrical and Electronics Engineers Inc., 2018 International Joint Conference on Neural Networks, IJCNN 2018, Rio de Janeiro, Brazil, 7/8/18. https://doi.org/10.1109/IJCNN.2018.8489393
Akcakaya M, Moeller S, Weingartner S, Ugurbil K. Subject-Specific Convolutional Neural Networks for Accelerated Magnetic Resonance Imaging. In 2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2018. 8489393. (Proceedings of the International Joint Conference on Neural Networks). https://doi.org/10.1109/IJCNN.2018.8489393
Akcakaya, Mehmet ; Moeller, Steen ; Weingartner, Sebastian ; Ugurbil, Kamil. / Subject-Specific Convolutional Neural Networks for Accelerated Magnetic Resonance Imaging. 2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. (Proceedings of the International Joint Conference on Neural Networks).
@inproceedings{66ddc438af9945fc8e87330678be5cae,
title = "Subject-Specific Convolutional Neural Networks for Accelerated Magnetic Resonance Imaging",
abstract = "Magnetic Resonance Imaging (MRI) is one of the leading modalities for medical imaging, providing excellent softtissue contrast without exposure to ionizing radiation. Despite continuing advances in MRI, long scan times remain a major limitation in clinical applications. Parallel imaging is a technique for scan time acceleration in MRI, which utilizes the spatial variations in the reception profiles of receiver coil arrays to reconstruct images from undersampled Fourier space, i.e. K-space. One of the most commonly used parallel imaging techniques employs interpolation of missing k-space information by using linear shiftinvariant convolutional kernels. These kernels are trained on a limited amount of autocalibration signal (ACS) for each scan. We propose a novel method for parallel imaging,Robust Artificialneural-networks for k -space Interpolation (RAKI), which uses scan-specific convolutional neural networks (CNNs) to perform improved k-space interpolation. Three-layer CNNs are trained using only scan-specific ACS data, alleviating the need for large training databases. The proposed method was tested in ultra-high resolution brain MRI and quantitative cardiac MRI, acquired with various acceleration rates. Improved noise resilience as compared to existing parallel imaging methods was observed for high acceleration rates or in the presence of low signal-tonoise ratio (SNR). Furthermore, RAKI successfully reconstructed images for quantitative cardiac MRI, even when using the same CNN across images with varying contrasts. These results indicate that RAKI achieves improved noise performance without overfitting to specific image contents, and offers great promise for improved acceleration in a wide range of MRI applications.",
author = "Mehmet Akcakaya and Steen Moeller and Sebastian Weingartner and Kamil Ugurbil",
year = "2018",
month = "10",
day = "10",
doi = "10.1109/IJCNN.2018.8489393",
language = "English (US)",
series = "Proceedings of the International Joint Conference on Neural Networks",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings",

}

TY - GEN

T1 - Subject-Specific Convolutional Neural Networks for Accelerated Magnetic Resonance Imaging

AU - Akcakaya, Mehmet

AU - Moeller, Steen

AU - Weingartner, Sebastian

AU - Ugurbil, Kamil

PY - 2018/10/10

Y1 - 2018/10/10

N2 - Magnetic Resonance Imaging (MRI) is one of the leading modalities for medical imaging, providing excellent softtissue contrast without exposure to ionizing radiation. Despite continuing advances in MRI, long scan times remain a major limitation in clinical applications. Parallel imaging is a technique for scan time acceleration in MRI, which utilizes the spatial variations in the reception profiles of receiver coil arrays to reconstruct images from undersampled Fourier space, i.e. K-space. One of the most commonly used parallel imaging techniques employs interpolation of missing k-space information by using linear shiftinvariant convolutional kernels. These kernels are trained on a limited amount of autocalibration signal (ACS) for each scan. We propose a novel method for parallel imaging,Robust Artificialneural-networks for k -space Interpolation (RAKI), which uses scan-specific convolutional neural networks (CNNs) to perform improved k-space interpolation. Three-layer CNNs are trained using only scan-specific ACS data, alleviating the need for large training databases. The proposed method was tested in ultra-high resolution brain MRI and quantitative cardiac MRI, acquired with various acceleration rates. Improved noise resilience as compared to existing parallel imaging methods was observed for high acceleration rates or in the presence of low signal-tonoise ratio (SNR). Furthermore, RAKI successfully reconstructed images for quantitative cardiac MRI, even when using the same CNN across images with varying contrasts. These results indicate that RAKI achieves improved noise performance without overfitting to specific image contents, and offers great promise for improved acceleration in a wide range of MRI applications.

AB - Magnetic Resonance Imaging (MRI) is one of the leading modalities for medical imaging, providing excellent softtissue contrast without exposure to ionizing radiation. Despite continuing advances in MRI, long scan times remain a major limitation in clinical applications. Parallel imaging is a technique for scan time acceleration in MRI, which utilizes the spatial variations in the reception profiles of receiver coil arrays to reconstruct images from undersampled Fourier space, i.e. K-space. One of the most commonly used parallel imaging techniques employs interpolation of missing k-space information by using linear shiftinvariant convolutional kernels. These kernels are trained on a limited amount of autocalibration signal (ACS) for each scan. We propose a novel method for parallel imaging,Robust Artificialneural-networks for k -space Interpolation (RAKI), which uses scan-specific convolutional neural networks (CNNs) to perform improved k-space interpolation. Three-layer CNNs are trained using only scan-specific ACS data, alleviating the need for large training databases. The proposed method was tested in ultra-high resolution brain MRI and quantitative cardiac MRI, acquired with various acceleration rates. Improved noise resilience as compared to existing parallel imaging methods was observed for high acceleration rates or in the presence of low signal-tonoise ratio (SNR). Furthermore, RAKI successfully reconstructed images for quantitative cardiac MRI, even when using the same CNN across images with varying contrasts. These results indicate that RAKI achieves improved noise performance without overfitting to specific image contents, and offers great promise for improved acceleration in a wide range of MRI applications.

UR - http://www.scopus.com/inward/record.url?scp=85056520301&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056520301&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2018.8489393

DO - 10.1109/IJCNN.2018.8489393

M3 - Conference contribution

T3 - Proceedings of the International Joint Conference on Neural Networks

BT - 2018 International Joint Conference on Neural Networks, IJCNN 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -