Dataset Information

Technical Note: Deep learning based MRAC using rapid ultrashort echo time imaging.

ABSTRACT:

Purpose

In this study, we explore the feasibility of a novel framework for MR-based attenuation correction for PET/MR imaging based on deep learning via convolutional neural networks, which enables fully automated and robust estimation of a pseudo CT image based on ultrashort echo time (UTE), fat, and water images obtained by a rapid MR acquisition.

Methods

MR images for MRAC are acquired using dual echo ramped hybrid encoding (dRHE), where both UTE and out-of-phase echo images are obtained within a short single acquisition (35 s). Tissue labeling of air, soft tissue, and bone in the UTE image is accomplished via a deep learning network that was pre-trained with T1-weighted MR images. UTE images are used as input to the network, which was trained using labels derived from co-registered CT images. The tissue labels estimated by deep learning are refined by a conditional random field based correction. The soft tissue labels are further separated into fat and water components using the two-point Dixon method. The estimated bone, air, fat, and water images are then assigned appropriate Hounsfield units, resulting in a pseudo CT image for PET attenuation correction. To evaluate the proposed MRAC method, PET/MR imaging of the head was performed on eight human subjects, where Dice similarity coefficients of the estimated tissue labels and relative PET errors were evaluated through comparison to a registered CT image.

Result

Dice coefficients for air (within the head), soft tissue, and bone labels were 0.76 ± 0.03, 0.96 ± 0.006, and 0.88 ± 0.01. In PET quantitation, the proposed MRAC method produced relative PET errors less than 1% within most brain regions.

Conclusion

The proposed MRAC method utilizing deep learning with transfer learning and an efficient dRHE acquisition enables reliable PET quantitation with accurate and rapid pseudo CT generation.

SUBMITTER: Jang H

PROVIDER: S-EPMC6443501 | biostudies-literature | 2018 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Technical Note: Deep learning based MRAC using rapid ultrashort echo time imaging.

Jang Hyungseok H Liu Fang F Zhao Gengyan G Bradshaw Tyler T McMillan Alan B AB

Medical physics 20180515

<h4>Purpose</h4>In this study, we explore the feasibility of a novel framework for MR-based attenuation correction for PET/MR imaging based on deep learning via convolutional neural networks, which enables fully automated and robust estimation of a pseudo CT image based on ultrashort echo time (UTE), fat, and water images obtained by a rapid MR acquisition.<h4>Methods</h4>MR images for MRAC are acquired using dual echo ramped hybrid encoding (dRHE), where both UTE and out-of-phase echo images ar ...[more]

PMID: 29763997

Similar Datasets

Project description:Background Signal contamination from long T2 water is a major challenge in direct imaging of myelin with MRI. Nulling of the unwanted long T2 signals can be achieved with an inversion recovery (IR) preparation pulse to null long T2 white matter within the brain. The remaining ultrashort T2 signal from myelin can be detected with an ultrashort echo time (UTE) sequence. Purpose To develop patient-specific whole-brain myelin imaging with a three-dimensional double-echo sliding inversion recovery (DESIRE) UTE sequence. Materials and Methods The DESIRE UTE sequence generates a series of IR images with different inversion times during a single scan. The optimal inversion time for nulling long T2 signal is determined by finding minimal signal on the second echo. Myelin images are generated by subtracting the second echo image from the first UTE image. To validate this method, a prospective study was performed in phantoms, cadaveric brain specimens, healthy volunteers, and patients with multiple sclerosis (MS). A total of 20 healthy volunteers (mean age, 40 years ± 13 [standard deviation], 10 women) and 20 patients with MS (mean age, 58 years ± 8; 15 women) who underwent MRI between November 2017 and February 2019 were prospectively included. Analysis of variance was performed to evaluate the signal difference between MS lesions and normal-appearing white matter in patients with MS. Results High signal intensity and corresponding T2* and T1 of the extracted myelin vesicles provided evidence for direct imaging of ultrashort-T2 myelin protons using the UTE sequence. Gadobenate dimeglumine phantoms with a wide range of T1 values were selectively suppressed with DESIRE UTE. In the ex vivo brain study of MS lesions, signal loss was observed in MS lesions and was conformed with histologic analysis. In the human study, there was a significant reduction in normalized signal intensity in MS lesions compared with that in normal-appearing white matter (0.19 ± 0.10 vs 0.76 ± 0.11, respectively; P < .001). Conclusion The double-echo sliding inversion recovery ultrashort echo time sequence can generate whole-brain myelin images specifically with a clinical 3-T scanner. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Port in this issue.

Project description:To assess the ability of ultrashort echo time (UTE) magnetic resonance (MR) imaging techniques to enable morphologic assessment of different types of meniscal calcifications, to compare these sequences with standard clinical sequences, and to perform T2* measurements of meniscal calcifications.This study was exempted by the institutional review board, and informed consent was not required. Ten human cadaveric menisci were imaged with high-spatial-resolution radiography and 3.0-T MR imaging by using morphologic (T1-weighted fast spin-echo [FSE], T2-weighted FSE, proton density [PD]-weighted FSE, two-dimensional [2D] fast spoiled gradient-echo [FSPGR], three-dimensional [3D] FSPGR, and 3D UTE) and quantitative (2D inversion-recovery [IR] UTE and 3D UTE) sequences. The menisci were divided into thirds for regional analysis. Morphologic assessment was performed with MR imaging; MR imaging findings were correlated with radiographs. Calcifications were classified as punctate, linear, or globular. T2* measurements were performed by manual placement of regions of interest (ROIs) in calcifications and by automatically creating ROIs in the surrounding tissues. Mixed-effects linear regression was used to determine variations in T2* as a function of region, morphology, and tissue type.The two globular calcifications were visualized with all sequences. For punctate (n=21) and linear (n=21) calcifications, respectively, visibility rates were as follows: 9.5% for both with the T1-weighted FSE sequence, 0% for both with the T2-weighted FSE sequence, 19.0% and 23.8% with the PD-weighted FSE sequence, 0% for both with the 2D IR UTE sequence, 100% for both with the 3D UTE sequence, and 100% for both with the 3D FSPGR sequence. T2* values were significantly lower for calcifications than for the surrounding meniscal tissue (P<.001). There was a trend of globular calcifications having lower T2* values than other morphologies (P=.08). With the 2D IR UTE technique, the T2* of the globular calcifications tended to be lower than with the 3D UTE technique (0.13-0.16 vs 1.32-3.03 msec) (P=.14, analysis of variance).UTE MR imaging sequences may allow morphologic as well as quantitative evaluation of meniscal calcifications.

Dataset Information

Technical Note: Deep learning based MRAC using rapid ultrashort echo time imaging.

Purpose

Methods

Result

Conclusion

Publications

Technical Note: Deep learning based MRAC using rapid ultrashort echo time imaging.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets