TY - JOUR
T1 - Super-Resolution 1H Magnetic Resonance Spectroscopic Imaging Utilizing Deep Learning
AU - Iqbal, Zohaib
AU - Nguyen, Dan
AU - Hangel, Gilbert
AU - Motyka, Stanislav
AU - Bogner, Wolfgang
AU - Jiang, Steve
N1 - Funding Information:
This research was supported by the research grants AP08857034 «Development of a new design of a pressing equipment and a chamber with a gas-dynamic installation with program control for the manufacture of additive technology of high-quality products» from the Ministry of Education and Science of the Republic of Kazakhstan for 2020-2022.»
Publisher Copyright:
© Copyright © 2019 Iqbal, Nguyen, Hangel, Motyka, Bogner and Jiang.
PY - 2019/10/9
Y1 - 2019/10/9
N2 - Magnetic resonance spectroscopic imaging (SI) is a unique imaging technique that provides biochemical information from in vivo tissues. The 1H spectra acquired from several spatial regions are quantified to yield metabolite concentrations reflective of tissue metabolism. However, since these metabolites are found in tissues at very low concentrations, SI is often acquired with limited spatial resolution. In this work, we test the hypothesis that deep learning is able to upscale low resolution SI, together with the T1-weighted (T1w) image, to reconstruct high resolution SI. We report on a novel densely connected UNet (D-UNet) architecture capable of producing super-resolution spectroscopic images. The inputs for the D-UNet are the T1w image and the low resolution SI image while the output is the high resolution SI. The results of the D-UNet are compared both qualitatively and quantitatively to simulated and in vivo high resolution SI. It is found that this deep learning approach can produce high quality spectroscopic images and reconstruct entire 1H spectra from low resolution acquisitions, which can greatly advance the current SI workflow.
AB - Magnetic resonance spectroscopic imaging (SI) is a unique imaging technique that provides biochemical information from in vivo tissues. The 1H spectra acquired from several spatial regions are quantified to yield metabolite concentrations reflective of tissue metabolism. However, since these metabolites are found in tissues at very low concentrations, SI is often acquired with limited spatial resolution. In this work, we test the hypothesis that deep learning is able to upscale low resolution SI, together with the T1-weighted (T1w) image, to reconstruct high resolution SI. We report on a novel densely connected UNet (D-UNet) architecture capable of producing super-resolution spectroscopic images. The inputs for the D-UNet are the T1w image and the low resolution SI image while the output is the high resolution SI. The results of the D-UNet are compared both qualitatively and quantitatively to simulated and in vivo high resolution SI. It is found that this deep learning approach can produce high quality spectroscopic images and reconstruct entire 1H spectra from low resolution acquisitions, which can greatly advance the current SI workflow.
KW - artificial intelligence
KW - deep learning (DL)
KW - magnetic resonance spectroscopic imaging (SI)
KW - magnetic resonance spectroscopy (H MRS)
KW - super-resolution
UR - http://www.scopus.com/inward/record.url?scp=85074168396&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85074168396&partnerID=8YFLogxK
U2 - 10.3389/fonc.2019.01010
DO - 10.3389/fonc.2019.01010
M3 - Article
C2 - 31649879
AN - SCOPUS:85074168396
VL - 9
JO - Frontiers in Oncology
JF - Frontiers in Oncology
SN - 2234-943X
M1 - 1010
ER -