Super-Resolution 1H Magnetic Resonance Spectroscopic Imaging Utilizing Deep Learning
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 metabol...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2019-10-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fonc.2019.01010/full |
| _version_ | 1811273703825080320 |
|---|---|
| author | Zohaib Iqbal Dan Nguyen Gilbert Hangel Stanislav Motyka Wolfgang Bogner Steve Jiang |
| author_facet | Zohaib Iqbal Dan Nguyen Gilbert Hangel Stanislav Motyka Wolfgang Bogner Steve Jiang |
| author_sort | Zohaib Iqbal |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-04-12T23:04:11Z |
| format | Article |
| id | doaj.art-f907c3cdc3234f06bccdaf24e8044b58 |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-04-12T23:04:11Z |
| publishDate | 2019-10-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-f907c3cdc3234f06bccdaf24e8044b582022-12-22T03:12:58ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2019-10-01910.3389/fonc.2019.01010465800Super-Resolution 1H Magnetic Resonance Spectroscopic Imaging Utilizing Deep LearningZohaib Iqbal0Dan Nguyen1Gilbert Hangel2Stanislav Motyka3Wolfgang Bogner4Steve Jiang5Medical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, TX, United StatesMedical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, TX, United StatesChristian Doppler Laboratory for Clinical Molecular MR Imaging, Department of Biomedical Imaging and Image-guided Therapy, High Field MR Center, Medical University of Vienna, Vienna, AustriaChristian Doppler Laboratory for Clinical Molecular MR Imaging, Department of Biomedical Imaging and Image-guided Therapy, High Field MR Center, Medical University of Vienna, Vienna, AustriaChristian Doppler Laboratory for Clinical Molecular MR Imaging, Department of Biomedical Imaging and Image-guided Therapy, High Field MR Center, Medical University of Vienna, Vienna, AustriaMedical Artificial Intelligence and Automation Laboratory, Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, TX, United StatesMagnetic 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.https://www.frontiersin.org/article/10.3389/fonc.2019.01010/fullsuper-resolutionmagnetic resonance spectroscopic imaging (SI)deep learning (DL)magnetic resonance spectroscopy (1H MRS)artificial intelligence |
| spellingShingle | Zohaib Iqbal Dan Nguyen Gilbert Hangel Stanislav Motyka Wolfgang Bogner Steve Jiang Super-Resolution 1H Magnetic Resonance Spectroscopic Imaging Utilizing Deep Learning Frontiers in Oncology super-resolution magnetic resonance spectroscopic imaging (SI) deep learning (DL) magnetic resonance spectroscopy (1H MRS) artificial intelligence |
| title | Super-Resolution 1H Magnetic Resonance Spectroscopic Imaging Utilizing Deep Learning |
| title_full | Super-Resolution 1H Magnetic Resonance Spectroscopic Imaging Utilizing Deep Learning |
| title_fullStr | Super-Resolution 1H Magnetic Resonance Spectroscopic Imaging Utilizing Deep Learning |
| title_full_unstemmed | Super-Resolution 1H Magnetic Resonance Spectroscopic Imaging Utilizing Deep Learning |
| title_short | Super-Resolution 1H Magnetic Resonance Spectroscopic Imaging Utilizing Deep Learning |
| title_sort | super resolution 1h magnetic resonance spectroscopic imaging utilizing deep learning |
| topic | super-resolution magnetic resonance spectroscopic imaging (SI) deep learning (DL) magnetic resonance spectroscopy (1H MRS) artificial intelligence |
| url | https://www.frontiersin.org/article/10.3389/fonc.2019.01010/full |
| work_keys_str_mv | AT zohaibiqbal superresolution1hmagneticresonancespectroscopicimagingutilizingdeeplearning AT dannguyen superresolution1hmagneticresonancespectroscopicimagingutilizingdeeplearning AT gilberthangel superresolution1hmagneticresonancespectroscopicimagingutilizingdeeplearning AT stanislavmotyka superresolution1hmagneticresonancespectroscopicimagingutilizingdeeplearning AT wolfgangbogner superresolution1hmagneticresonancespectroscopicimagingutilizingdeeplearning AT stevejiang superresolution1hmagneticresonancespectroscopicimagingutilizingdeeplearning |