Golden-angle Radial Sparse Parallel Reconstruction Research Articles

Ultra-rapid, Free-breathing, Real-time Cardiac Cine MRI Using GRASP Amplified with View Sharing and KWIC Filtering.

Purpose To achieve ultra-high temporal resolution (approximately 20 msec) in free-breathing, real-time cardiac cine MRI using golden-angle radial sparse parallel (GRASP) reconstruction amplified with view sharing (VS) and k-space-weighted image contrast (KWIC) filtering. Materials and Methods Fourteen pediatric patients with congenital heart disease (mean age [SD], 9 years ± 2; 13 male) and 10 adult patients with arrhythmia (mean age, 62 years ± 8; nine male) who underwent both standard breath-hold cine and free-breathing real-time cine using GRASP were retrospectively identified. To achieve high temporal resolution, each time frame was reconstructed using six radial spokes, corresponding to acceleration factors ranging from 24 to 32. To compensate for loss in spatial resolution resulting from over-regularization in GRASP, VS and KWIC filtering were incorporated. The blur metric, visual image quality scores, and biventricular parameters were compared between clinical and real-time cine images. Results In pediatric patients, the incorporation of VS and KWIC into GRASP (ie, GRASP + VS + KWIC) produced significantly (P < .05) sharper x-y-t (blur metric: 0.36 ± 0.03, 0.41 ± 0.03, 0.48 ± 0.03, respectively) and x-y-f (blur metric: 0.28 ± 0.02, 0.31 ± 0.03, 0.37 ± 0.03, respectively) component images compared with GRASP + VS and conventional GRASP. Only the noise score differed significantly between GRASP + VS + KWIC and clinical cine; all visual scores were above the clinically acceptable (3.0) cutoff point. Biventricular volumetric parameters strongly correlated (R2 > 0.85) between clinical and real-time cine images reconstructed with GRASP + VS + KWIC and were in good agreement (relative error < 6% for all parameters). In adult patients, the visual scores of all categories were significantly lower (P < .05) for clinical cine compared with real-time cine with GRASP + VS + KWIC, except for noise (P = .08). Conclusion Incorporating VS and KWIC filtering into GRASP reconstruction enables ultra-high temporal resolution (approximately 20 msec) without significant loss in spatial resolution. Keywords: Cine, View Sharing, k-Space-weighted Image Contrast Filtering, Radial k-Space, Pediatrics, Arrhythmia, GRASP, Compressed Sensing, Real-Time, Free-Breathing Supplemental material is available for this article. © RSNA, 2024.

Correcting versus resolving respiratory motion in free-breathing whole-heart MRA: a comparison in patients with thoracic aortic disease

BackgroundWhole-heart magnetic resonance angiography (MRA) requires sophisticated methods accounting for respiratory motion. Our purpose was to evaluate the image quality of compressed sensing-based respiratory motion-resolved three-dimensional (3D) whole-heart MRA compared with self-navigated motion-corrected whole-heart MRA in patients with known thoracic aorta dilation.MethodsTwenty-five patients were prospectively enrolled in this ethically approved study. Whole-heart 1.5-T MRA was acquired using a prototype 3D radial steady-state free-precession free-breathing sequence. The same data were reconstructed with a one-dimensional motion-correction algorithm (1D-MCA) and an extradimensional golden-angle radial sparse parallel reconstruction (XD-GRASP). Subjective image quality was scored and objective image quality was quantified (signal intensity ratio, SIR; vessel sharpness). Wilcoxon, McNemar, and paired t tests were used.ResultsSubjective image quality was significantly higher using XD-GRASP compared to 1D-MCA (median 4.5, interquartile range 4.5–5.0 versus 4.0 [2.25–4.75]; p < 0.001), as well as signal homogeneity (3.0 [3.0–3.0] versus 2.0 [2.0–3.0]; p = 0.003), and image sharpness (3.0 [2.0–3.0] vs 2.0 [1.25–3.0]; p < 0.001). SIR with the 1D-MCA and XD-GRASP was 6.1 ± 3.9 versus 7.4 ± 2.5, respectively (p < 0.001); while signal homogeneity was 274.2 ± 265.0 versus 199.8 ± 67.2 (p = 0.129). XD-GRASP provided a higher vessel sharpness (45.3 ± 10.7 versus 40.6 ± 101, p = 0.025).ConclusionsXD-GRASP-based motion-resolved reconstruction of free-breathing 3D whole-heart MRA datasets provides improved image contrast, sharpness, and signal homogeneity and seems to be a promising technique that overcomes some of the limitations of motion correction or respiratory navigator gating.

Does GRASP affect DCE-MRI quantitative parameters and texture analysis in patients with esophageal cancer receiving preoperative neoadjuvant chemotherapy?

To compare pharmacokinetic parameters and texture features from two dynamic contrast-enhanced (DCE) MR images [golden-angle radial sparse parallel (GRASP) and view-sharing with golden-angle radial profile (VS-GR) reconstruction], and explore their value in assessing response to neoadjuvant chemotherapy (nCT) in esophageal cancer (EC). The study prospectively enrolled 30 EC patients receiving nCT before surgery. DCE-MRI scanning was performed before nCT and within 1 week before surgery. Chemotherapy response was assessed according to RECIST 1.1 and Tumor Regression Grade (TRG). Mann–Whitney U test was utilized for comparing GRASP and VS-GR reconstruction, and the receiver operating characteristic (ROC) was performed for each significant feature to assess its accuracy in predicting response. Among the 30 patients included in this cohort (28 men; average age of 58 ± 8 years), response by RECIST 1.1 demonstrated 18 responders and 12 non-responders. For TRG, no case showed TRG1, 1 patient was TRG2, 3 patients were TRG3, 8 patients were TRG4, and 18 patients were TRG5. A total of 72 pharmacokinetic parameters and texture features were extracted from each tumor. Of those, 29 pre-nCT features and 24 post-nCT features showed statistically significant difference between GRASP and VS-GR reconstruction. One pre-nCT texture feature and 37 post-nCT pharmacokinetic parameters and texture features on VS-GR showed statistically significant differences between responder and non-responders. Both pre- and post-nCT pharmacokinetic parameters and texture features with GRASP reconstruction showed good performance in response groups (AUC > 0.70, P 0.70, P < 0.05). GRASP can improve pharmacokinetic parameters and texture features from DCE-MR imaging. Both pre- and post-nCT pharmacokinetic parameters and texture features with GRASP and only post-nCT those with VS-GR reconstruction showed the ability to assess the response to nCT in EC.

Golden-Angle Radial Sparse Parallel MR Image Reconstruction Using SC-GROG Followed by Iterative Soft Thresholding

Golden-angle radial sparse parallel (GRASP) magnetic resonance imaging (MRI) is a recent MR image reconstruction technique which integrates parallel imaging, compressed sensing and golden-angle radial scheme to reconstruct the dynamic contrast-enhanced MRI (DCE-MRI) data. Conventionally, GRASP exploits non-uniform fast Fourier transform to grid and de-grid the golden-angle radial data and employs nonlinear conjugate gradient method to recover the unaliased images. GRASP performs gridding and de-gridding operations of golden-angle radial data in every iteration which increases the computational complexity of the conventional GRASP and takes a long image reconstruction time. In this paper, self-calibrated GRAPPA operator gridding (SC-GROG) followed by iterative soft thresholding (IST) is proposed for faster GRASP reconstruction of the golden-angle radial DCE-MRI data. In the proposed method, firstly SC-GROG maps the undersampled golden-angle radial data to a Cartesian grid and then reconstructs the solution image using the IST technique. The proposed method does not require gridding and de-gridding in each iteration; therefore, it is computationally less expensive as compared to the conventional GRASP reconstruction approach. The proposed method is tested for undersampled DCE golden-angle radial liver perfusion data (at acceleration factors 11.8, 19.1 and 30.9). The reconstruction results are assessed visually as well as using mean square error, line profiles and reconstruction time. The reconstruction results are compared with the conventional GRASP reconstruction. The results show that the proposed method provides better quality reconstruction results in terms of reconstruction time and spatio-temporal resolution than the conventional GRASP approach.

Free-breathing dynamic contrast-enhanced MRI for assessment of pulmonary lesions using golden-angle radial sparse parallel imaging.

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has been shown to be a promising technique for assessing lung lesions. However, DCE-MRI often suffers from motion artifacts and insufficient imaging speed. Therefore, highly accelerated free-breathing DCE-MRI is of clinical interest for lung exams. To test the performance of rapid free-breathing DCE-MRI for simultaneous qualitative and quantitative assessment of pulmonary lesions using Golden-angle RAdial Sparse Parallel (GRASP) imaging. Prospective. Twenty-six patients (17 males, mean age = 55.1 ± 14.4) with known pulmonary lesions. 3T MR scanner; a prototype fat-saturated, T1 -weighted stack-of-stars golden-angle radial sequence for data acquisition and a Cartesian breath-hold volumetric-interpolated examination (BH-VIBE) sequence for comparison. After a dual-mode GRASP reconstruction, one with 3-second temporal resolution (3s-GRASP) and the other with 15-second temporal resolution (15s-GRASP), all GRASP and BH-VIBE images were pooled together for blind assessment by two experienced radiologists, who independently scored the overall image quality, lesion delineation, overall artifact level, and diagnostic confidence of each case. Perfusion analysis was performed for the 3s-GRASP images using a Tofts model to generate the volume transfer coefficient (Ktrans ) and interstitial volume (Ve ). Nonparametric paired two-tailed Wilcoxon signed-rank test; Cohen's kappa; unpaired Student's t-test. 15s-GRASP achieved comparable image quality with conventional BH-VIBE (P > 0.05), except for the higher overall artifact level in the precontrast phase (P = 0.018). The Ktrans and Ve in inflammation were higher than those in malignant lesions (Ktrans : 0.78 ± 0.52 min-1 vs. 0.37 ± 0.22 min-1 , P = 0.020; Ve : 0.36 ± 0.16 vs. 0.26 ± 0.1, P = 0.177). Also, the Ktrans and Ve in malignant lesions were also higher than those in benign lesions (Ktrans : 0.37 ± 0.22 min-1 vs. 0.04 ± 0.04 min-1 , P = 0.001; Ve : 0.26 ± 0.12 vs. 0.10 ± 0.00, P = 0.063). This feasibility study demonstrated the performance of high spatiotemporal resolution free-breathing DCE-MRI of the lung using GRASP for qualitative and quantitative assessment of pulmonary lesions. 2 Technical Efficacy: Stage 1 J. MAGN. RESON. IMAGING 2018;48:459-468.

Optimization and validation of accelerated golden-angle radial sparse MRI reconstruction with self-calibrating GRAPPA operator gridding.

Golden-angle radial sparse parallel (GRASP) MRI reconstruction requires gridding and regridding to transform data between radial and Cartesian k-space. These operations are repeatedly performed in each iteration, which makes the reconstruction computationally demanding. This work aimed to accelerate GRASP reconstruction using self-calibrating GRAPPA operator gridding (GROG) and to validate its performance in clinical imaging. GROG is an alternative gridding approach based on parallel imaging, in which k-space data acquired on a non-Cartesian grid are shifted onto a Cartesian k-space grid using information from multicoil arrays. For iterative non-Cartesian image reconstruction, GROG is performed only once as a preprocessing step. Therefore, the subsequent iterative reconstruction can be performed directly in Cartesian space, which significantly reduces computational burden. Here, a framework combining GROG with GRASP (GROG-GRASP) is first optimized and then compared with standard GRASP reconstruction in 22 prostate patients. GROG-GRASP achieved approximately 4.2-fold reduction in reconstruction time compared with GRASP (∼333 min versus ∼78 min) while maintaining image quality (structural similarity index ≈ 0.97 and root mean square error ≈ 0.007). Visual image quality assessment by two experienced radiologists did not show significant differences between the two reconstruction schemes. With a graphics processing unit implementation, image reconstruction time can be further reduced to approximately 14 min. The GRASP reconstruction can be substantially accelerated using GROG. This framework is promising toward broader clinical application of GRASP and other iterative non-Cartesian reconstruction methods. Magn Reson Med 80:286-293, 2018. © 2017 International Society for Magnetic Resonance in Medicine.

Role of High-Resolution Dynamic Contrast-Enhanced MRI with Golden-Angle Radial Sparse Parallel Reconstruction to Identify the Normal Pituitary Gland in Patients with Macroadenomas.

Preoperative localization of the pituitary gland with imaging in patients with macroadenomas has been inadequately explored. The pituitary gland enhancing more avidly than a macroadenoma has been described in the literature. Taking advantage of this differential enhancement pattern, our aim was to evaluate the role of high-resolution dynamic MR imaging with golden-angle radial sparse parallel reconstruction in localizing the pituitary gland in patients undergoing trans-sphenoidal resection of a macroadenoma. A retrospective study was performed in 17 patients who underwent trans-sphenoidal surgery for pituitary macroadenoma. Radial volumetric interpolated brain examination sequences with golden-angle radial sparse parallel technique were obtained. Using an ROI-based method to obtain signal-time curves and permeability measures, 3 separate readers identified the normal pituitary gland distinct from the macroadenoma. The readers' localizations were then compared with the intraoperative location of the gland. Statistical analyses were performed to assess the interobserver agreement and correlation with operative findings. The normal pituitary gland was found to have steeper enhancement-time curves as well as higher peak enhancement values compared with the macroadenoma (P < .001). Interobserver agreement was almost perfect in all 3 planes (κ = 0.89). In the 14 cases in which the gland was clearly identified intraoperatively, the correlation between the readers' localization and the true location derived from surgery was also nearly perfect (κ = 0.95). This study confirms our ability to consistently and accurately identify the normal pituitary gland in patients with macroadenomas with the golden-angle radial sparse parallel technique with quantitative permeability measurements and enhancement-time curves.

Respiratory self-gated 3D UTE for lung imaging in small animal MRI.

To investigate retrospective respiratory gating of three-dimensional ultrashort echo time (3D UTE) lung acquisition in free-breathing rats using k-space center self gating signal (DC-SG) and 3D image-based SG (3D-Img-SG). Seven rats were investigated with a quasi-random 3D UTE protocol. Low-resolution time-resolved sliding-window images were reconstructed with a 3D golden-angle radial sparse parallel (GRASP) reconstruction to extract a 3D-Img-SG signal, whereas DC-SG was extracted from the center of k-space. Both signals were sorted into 10 respiratory bins. Signal-to-noise ratio (SNR) and normalized signal intensity (NSI) in lung parenchyma, image sharpness, and lung volume changes were studied in the resulting images to show feasibility of the method. An algorithm for bulk movement identification and removal was implemented. Three-dimensional Img-SG allows reconstruction of different respiratory stages in all acquired datasets, showing clear differences in diaphragm position and significantly different lung volumes, SNR, and NSI in lung parenchyma. Improved sharpness in expiration images was observed compared to ungated images. DC-SG did not result in clear different diaphragm position in all cases. Bulk motion removal improved final image sharpness. Low-resolution 3D GRASP reconstruction allowed for extraction of an effective gating signal for 3D-Img-SG. The DC-SG method did not work in cases for which respiratory frequencies were inconsistent. Magn Reson Med 78:739-745, 2017. © 2016 International Society for Magnetic Resonance in Medicine.