scholarly journals Boosting the Signal-to-Noise of Low-Field MRI With Deep Learning Image Reconstruction

2020 ◽  
Author(s):  
Neha Koonjoo ◽  
Bo Zhu ◽  
Cody Bagnall ◽  
Matthew Rosen
Keyword(s):  
Image Reconstruction ◽  
Image Quality ◽  
Plant Root ◽  
Signal Acquisition ◽  
Neural Network Approach ◽  
Low Field ◽  
Reconstruction Performance ◽  
Low Field Mri ◽  
Brain Data ◽  
The Impact

Abstract Recent years have seen a resurgence of interest in inexpensive low-field (<0.3 T) MRI systems mainly due to advances in magnet, coil and gradient set designs. However, most of these advances are focused on hardware development and signal acquisition while far less attention has been given to how advanced image reconstruction can improve image quality at low field. We describe here the use of our end-to-end deep neural network approach (AUTOMAP) to improve the image quality of highly noise-corrupted low-field MRI data. We compare the performance of this approach to two additional state-of-the-art denoising pipelines. We find that AUTOMAP improves image reconstruction of data acquired on two very different low-field MRI systems: human brain data acquired at 6.5 mT, and plant root data acquired at 47 mT, demonstrating SNR gains above Fourier reconstruction by factors of 1.5- to 4.5-fold, and 3-fold, respectively. In these applications, AUTOMAP outperformed both contemporary denoising algorithms and suppressed noise-like spike artifacts in reconstructed images. The impact of domain-specific training corpora on the reconstruction performance is discussed. The AUTOMAP approach to image reconstruction will enable significant image quality improvements at low-field, especially in highly noise-corrupted environments.

scholarly journals Boosting the signal-to-noise of low-field MRI with deep learning image reconstruction

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
N. Koonjoo ◽  
B. Zhu ◽  
G. Cody Bagnall ◽  
D. Bhutto ◽  
M. S. Rosen
Keyword(s):  
Image Reconstruction ◽  
Image Quality ◽  
Plant Root ◽  
Signal Acquisition ◽  
Neural Network Approach ◽  
Low Field ◽  
Reconstruction Methods ◽  
Reconstruction Performance ◽  
Low Field Mri ◽  
The Impact

AbstractRecent years have seen a resurgence of interest in inexpensive low magnetic field (< 0.3 T) MRI systems mainly due to advances in magnet, coil and gradient set designs. Most of these advances have focused on improving hardware and signal acquisition strategies, and far less on the use of advanced image reconstruction methods to improve attainable image quality at low field. We describe here the use of our end-to-end deep neural network approach (AUTOMAP) to improve the image quality of highly noise-corrupted low-field MRI data. We compare the performance of this approach to two additional state-of-the-art denoising pipelines. We find that AUTOMAP improves image reconstruction of data acquired on two very different low-field MRI systems: human brain data acquired at 6.5 mT, and plant root data acquired at 47 mT, demonstrating SNR gains above Fourier reconstruction by factors of 1.5- to 4.5-fold, and 3-fold, respectively. In these applications, AUTOMAP outperformed two different contemporary image-based denoising algorithms, and suppressed noise-like spike artifacts in the reconstructed images. The impact of domain-specific training corpora on the reconstruction performance is discussed. The AUTOMAP approach to image reconstruction will enable significant image quality improvements at low-field, especially in highly noise-corrupted environments.

Improving Image Quality In Low-Field MRI With Deep Learning

2021 ◽  
Author(s):  
Armando Garcia Hernandez ◽  
Pierre Fau ◽  
Stanislas Rapacchi ◽  
Julien Wojak ◽  
Hugues Mailleux ◽  
...  
Keyword(s):  
Deep Learning ◽  
Image Quality ◽  
Low Field ◽  
Low Field Mri

scholarly journals Image quality assessment of deep learning image reconstruction in computed tomography using tube current modulation: a phantom study

2021 ◽  
Author(s):  
kazuhiro takeuchi ◽  
Yasuhiro Ide ◽  
Yuichiro Mori ◽  
Yusuke Uehara ◽  
Hiroshi Sukeishi ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Deep Learning ◽  
Image Reconstruction ◽  
Image Quality ◽  
Quality Characteristics ◽  
Image Noise ◽  
Tube Current Modulation ◽  
Current Modulation ◽  
Tube Current ◽  
The Impact

Abstract The novel deep learning image reconstruction (DLIR) is known to change its image quality characteristics according to object contrast and image noise. In clinical practice, computed tomography (CT) image noise is usually controlled by tube current modulation (TCM) to accommodate changes in object size. This study aimed to evaluate the image quality characteristics of DLIR for different object sizes when in-plane noise is controlled by TCM. We used Mercury 4.0 phantoms with different object sizes. Phantom image acquisition was performed on a GE Revolution CT system to investigate the impact of the DLIR algorithm compared to standard reconstructions: filtered back projection (FBP) and hybrid iterative reconstruction (hybrid-IR). For image quality evaluation, the noise power spectrum (NPS), task-based transfer function (TTF), and detectability index (d') were determined. The NPS of DLIR was very similar to that of FBP, and the information in the high-frequency region was maintained. In terms of TTF, DLIR showed higher resolution than hybrid-IR at low- to medium-contrast (Δ50, Δ90HU), but not necessarily higher than FBP. At the simulated contrast and lesion size, DLIR showed higher detectability than hybrid-IR, regardless of the phantom size. In this study, we evaluated a novel DLIR algorithm by reproducing clinical behaviors. The findings indicate that DLIR produces higher image quality than hybrid-IR regardless of the phantom size, although it depends on the reconstruction strength.

scholarly journals Approaches for Image Reconstruction in Low-Field Magnetic Resonance Imaging

2021 ◽  
Author(s):  
Johnes Obungoloch ◽  
Emmanuel Ahishakiye
Keyword(s):  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Image Reconstruction ◽  
Data Driven ◽  
Resonance Imaging ◽  
Learning Methods ◽  
Image Reconstruction Techniques ◽  
Low Field ◽  
Low Field Mri ◽  
Magnetic Resonance Imaging Mri

Abstract Background: Magnetic Resonance Imaging (MRI) and spectroscopic techniques are frequently employed for clinical diagnostics as well as basic research in areas like cognitive neuroimaging. MRI is a widely used imaging modality for intracranial diseases. However, conventional MRI is expensive to purchase, maintain and sustain, limiting their use in low-income countries. Low field MRI can provide an economical, long-term, and safe imaging option to high-field MRI and computed tomography (CT) for brain imaging. This paper offers a review of the image reconstruction techniques used in low field magnetic resonance imaging (MRI). It is aimed at familiarizing the readers with the relevant knowledge, literature, and the latest updates on the state-of-art image reconstruction techniques that have been used in low field MRI citing their strengths, and areas for improvement. Methods: An in-depth keyword-based search was undertaken for publications on image reconstruction approaches in low-field MRI in the top scientific databases such as Google Scholar, Wiley, Science Direct, Springer, IEEE, Scopus, Nature, Elsevier, and PubMed throughout this study. This research also contained relevant postgraduate theses. For the selection of relevant research publications, the PRISMA flow diagram and protocol were also used.Results: Studies revealed that Inhomogeneities are present in low field MRI, implying that the traditional method of acquiring the image, using the inverse Fourier Transform, is no longer viable. The image reconstruction techniques reviewed include iterative methods, dictionary learning methods, and deep learning methods. Experimental results from the literature revealed improved image quality of the reconstructed images using data driven and learning based methods (deep learning and dictionary learning methods). Conclusion: The study revealed that there is limited literature on the image reconstruction approaches in low field MRI even if though there are sufficient studies on the subject in high field MRI. Data driven and learning based methods improves image reconstruction quality when compared to analytic and iterative approaches.

High-Quality MicroCT Rock Imaging: Methodology To Measure and Correct for X-Ray Scatter

SPE Journal ◽  
2019 ◽  
Vol 25 (01) ◽  
pp. 226-241
Author(s):  
Alexander Katsevich ◽  
Michael Frenkel ◽  
Qiushi Sun ◽  
Shannon L. Eichmann ◽  
Victor Prieto
Keyword(s):  
Image Reconstruction ◽  
Image Quality ◽  
Scatter Correction ◽  
The United States ◽  
Accurate Estimation ◽  
High Quality ◽  
X Ray ◽  
The Core ◽  
Ct Data ◽  
The Impact

Summary Microcomputed tomography (microCT) of cores yields valuable information about rock and fluid properties at pore scale for conventional rock and at rock heterogeneity scale for unconventionals. High levels of uncorrected X-ray scatter in computed tomography (CT) data lead to strong image artifacts and erroneous Hounsfield unit (HU) values, making reconstructed images unsuitable for accurate digital rock (DR) characterization (e.g., segmentation, material decomposition, and others). MicroCT scanners do not include scatter correction techniques. To fill this gap, we developed a new methodology to measure and remove the scatter component from raw projection microCT data collected during rock core scans, and ultimately improving the image quality of scanned cores. Widely used approaches for scatter estimation, based on Monte Carlo (MC) simulations and simplified analytical models, are time-consuming and may lose accuracy when imaging complex unconventional shale cores. In this paper, we propose a more practical approach to perform scatter correction from direct scatter measurements, an approach that is based on the beam-stop array (BSA) method. The BSA method works as follows: The radiation scattered by the core sample is emitted in random directions. By placing an array of small, highly absorbing beads between the source and the core, the primary X-ray signal through the beads is blocked, but the overall object scatter signal is not affected. The observed values in the beads’ shadows on the detector are assumed to be scatter signal. Performing interpolation of the scatter signal between the shadowed by beads pixels on the detector gives an estimate of the scatter signal at every pixel on the detector. Subtracting scatter from projection data yields scatter-corrected data used for 3D CT core image reconstruction. To develop the core scatter correction methodology, we executed the following three tasks: (1) performed modeling of primary and scattered signals to optimize the BSA design (beads size and layout) and scan parameters; (2) developed and implemented an accurate scatter correction algorithm into our 3D microCT image reconstruction workflow; and (3) tested the proposed methodology using four shale core samples from the United States and the Middle East. To better assess the impact of scatter, all experiments with shale core plugs presented in this paper were conducted using source energy of 160 kVp. Our results demonstrated that in many cases, especially with higher attenuating cores, failing to correct for X-ray scatter may result in significant loss of image reconstruction accuracy. We also showed that the developed methodology allows for accurate estimation and removal of scatter from the raw (projection) CT data, enabling reconstruction of high-quality core images that are required for performing DR analysis. To assess the impact of X-ray scatter on the accuracy of DR segmentation, we compared the amount of resolved air-filled space using a stack of image slices by thresholding for the air regions. Our results showed that the amount of detected air-filled space may increase significantly when scatter correction is applied. The presented scatter correction methodology is general and can be used with any microCT scanner used by the petroleum industry to improve image quality and derive accurate HU values. This is of significant importance for quantitative characterization of highly heterogeneous rock with fine structural changes, as is the case for shale. Ultimately, this methodology should expand the operational envelope and value of microCT imaging in the exploration and production workflows.

scholarly journals Accelerated T2-Weighted TSE Imaging of the Prostate Using Deep Learning Image Reconstruction: A Prospective Comparison with Standard T2-Weighted TSE Imaging

Cancers ◽  
2021 ◽  
Vol 13 (14) ◽  
pp. 3593
Author(s):  
Sebastian Gassenmaier ◽  
Saif Afat ◽  
Marcel Dominik Nickel ◽  
Mahmoud Mostapha ◽  
Judith Herrmann ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Deep Learning ◽  
Image Reconstruction ◽  
Image Quality ◽  
Acquisition Time ◽  
Noise Levels ◽  
Diagnostic Confidence ◽  
Prospective Comparison ◽  
The Impact ◽  
Lesion Conspicuity

Multiparametric MRI (mpMRI) of the prostate has become the standard of care in prostate cancer evaluation. Recently, deep learning image reconstruction (DLR) methods have been introduced with promising results regarding scan acceleration. Therefore, the aim of this study was to investigate the impact of deep learning image reconstruction (DLR) in a shortened acquisition process of T2-weighted TSE imaging, regarding the image quality and diagnostic confidence, as well as PI-RADS and T2 scoring, as compared to standard T2 TSE imaging. Sixty patients undergoing 3T mpMRI for the evaluation of prostate cancer were prospectively enrolled in this institutional review board-approved study between October 2020 and March 2021. After the acquisition of standard T2 TSE imaging (T2S), the novel T2 TSE sequence with DLR (T2DLR) was applied in three planes. Overall, the acquisition time for T2S resulted in 10:21 min versus 3:50 min for T2DLR. The image evaluation was performed by two radiologists independently using a Likert scale ranging from 1–4 (4 best) applying the following criteria: noise levels, artifacts, overall image quality, diagnostic confidence, and lesion conspicuity. Additionally, T2 and PI-RADS scoring were performed. The mean patient age was 69 ± 9 years (range, 49–85 years). The noise levels and the extent of the artifacts were evaluated to be significantly improved in T2DLR versus T2S by both readers (p < 0.05). Overall image quality was also evaluated to be superior in T2DLR versus T2S in all three acquisition planes (p = 0.005–<0.001). Both readers evaluated the item lesion conspicuity to be superior in T2DLR with a median of 4 versus a median of 3 in T2S (p = 0.001 and <0.001, respectively). T2-weighted TSE imaging of the prostate in three planes with an acquisition time reduction of more than 60% including DLR is feasible with a significant improvement of image quality.

Improved Contrast in Ultra-Low-Field MRI with Time-Dependent Bipolar Prepolarizing Fields: Theory and NMR Demonstrations

2013 ◽  
Vol 20 (3) ◽  
pp. 327-336 ◽  
Author(s):  
Jaakko O. Nieminen ◽  
Jens Voigt ◽  
Stefan Hartwig ◽  
Hans Jürgen Scheer ◽  
Martin Burghoff ◽  
...  
Keyword(s):  
Field Strength ◽  
Time Course ◽  
Signal Strength ◽  
Relaxation Rates ◽  
Resonance Imaging ◽  
New Approach ◽  
Spin Lattice ◽  
Low Field ◽  
Low Field Mri ◽  
Magnetic Resonance Imaging Mri

Abstract The spin-lattice (T1) relaxation rates of materials depend on the strength of the external magnetic field in which the relaxation occurs. This T1 dispersion has been suggested to offer a means to discriminate between healthy and cancerous tissue by performing magnetic resonance imaging (MRI) at low magnetic fields. In prepolarized ultra-low-field (ULF) MRI, spin precession is detected in fields of the order of 10-100 μT. To increase the signal strength, the sample is first magnetized with a relatively strong polarizing field. Typically, the polarizing field is kept constant during the polarization period. However, in ULF MRI, the polarizing-field strength can be easily varied to produce a desired time course. This paper describes how a novel variation of the polarizing-field strength and duration can optimize the contrast between two types of tissue having different T1 relaxation dispersions. In addition, NMR experiments showing that the principle works in practice are presented. The described procedure may become a key component for a promising new approach of MRI at ultra-low fields

scholarly journals P‐13.5: The impact of Gate Loading on Image Quality for large & middle size AM‐OLED display devices

2021 ◽  
Vol 52 (S1) ◽  
pp. 643-646
Author(s):  
Yang Guobo ◽  
Qiu Haijun ◽  
Huang Weiyun ◽  
Yang Yuqing ◽  
Long Yue ◽  
...  
Keyword(s):  
Image Quality ◽  
Display Devices ◽  
The Impact ◽  
Oled Display

scholarly journals Constrained Backtracking Matching Pursuit Algorithm for Image Reconstruction in Compressed Sensing

2021 ◽  
Vol 11 (4) ◽  
pp. 1435
Author(s):  
Xue Bi ◽  
Lu Leng ◽  
Cheonshik Kim ◽  
Xinwen Liu ◽  
Yajun Du ◽  
...  
Keyword(s):  
Image Reconstruction ◽  
Compressed Sensing ◽  
Matching Pursuit ◽  
Reconstruction Algorithm ◽  
Energy Criterion ◽  
Step Size ◽  
Relationship Analysis ◽  
Support Set ◽  
Reconstruction Performance ◽  
Sparsity Level

Image reconstruction based on sparse constraints is an important research topic in compressed sensing. Sparsity adaptive matching pursuit (SAMP) is a greedy pursuit reconstruction algorithm, which reconstructs signals without prior information of the sparsity level and potentially presents better reconstruction performance than other greedy pursuit algorithms. However, SAMP still suffers from being sensitive to the step size selection at high sub-sampling ratios. To solve this problem, this paper proposes a constrained backtracking matching pursuit (CBMP) algorithm for image reconstruction. The composite strategy, including two kinds of constraints, effectively controls the increment of the estimated sparsity level at different stages and accurately estimates the true support set of images. Based on the relationship analysis between the signal and measurement, an energy criterion is also proposed as a constraint. At the same time, the four-to-one rule is improved as an extra constraint. Comprehensive experimental results demonstrate that the proposed CBMP yields better performance and further stability than other greedy pursuit algorithms for image reconstruction.

Sign in / Sign up

Export Citation Format

Share Document