Prospective Deployment of Deep Learning Reconstruction Facilitates Highly Accelerated Upper Abdominal MRI.

RATIONALE AND OBJECTIVES: To compare a conventional T1 volumetric interpolated breath-hold examination (VIBE) with SPectral Attenuated Inversion Recovery (SPAIR) fat saturation and a deep learning (DL)-reconstructed accelerated VIBE sequence with SPAIR fat saturation achieving a 50 % reduction in breath-hold duration (hereafter, VIBE-SPAIRDL) in terms of image quality and diagnostic confidence. MATERIALS AND METHODS: This prospective study enrolled consecutive patients referred for upper abdominal MRI from November 2023 to December 2023 at a single tertiary center. Patients underwent upper abdominal MRI with acquisition of non-contrast and gadobutrol-enhanced conventional VIBE-SPAIR (fourfold acceleration, acquisition time 16 s) and VIBE-SPAIRDL (sixfold acceleration, acquisition time 8 s) on a 1.5 T scanner. Image analysis was performed by four readers, evaluating homogeneity of fat suppression, perceived signal-to-noise ratio (SNR), edge sharpness, artifact level, lesion detectability and diagnostic confidence. A statistical power analysis for patient sample size estimation was performed. Image quality parameters were compared by a repeated measures analysis of variance, and interreader agreement was assessed using Fleiss' κ. RESULTS: Among 450 consecutive patients, 45 patients were evaluated (mean age, 60 years ± 15 [SD]; 27 men, 18 women). VIBE-SPAIRDL acquisition demonstrated superior SNR (P < 0.001), edge sharpness (P < 0.001), and reduced artifacts (P < 0.001) with substantial to almost perfect interreader agreement for non-contrast (κ: 0.70-0.91) and gadobutrol-enhanced MRI (κ: 0.68-0.87). No evidence of a difference was found between conventional VIBE-SPAIR and VIBE-SPAIRDL regarding homogeneity of fat suppression, lesion detectability, or diagnostic confidence (all P > 0.05). CONCLUSION: Deep learning reconstruction of VIBE-SPAIR facilitated a reduction of breath-hold duration by half, while reducing artifacts and improving image quality. SUMMARY: Deep learning reconstruction of prospectively accelerated T1 volumetric interpolated breath-hold examination for upper abdominal MRI enabled a 50 % reduction in breath-hold time with superior image quality. KEY RESULTS: 1) In a prospective analysis of 45 patients referred for upper abdominal MRI, accelerated deep learning (DL)-reconstructed VIBE images with spectral fat saturation (SPAIR) showed better overall image quality, with better perceived signal-to-noise ratio and less artifacts (all P < 0.001), despite a 50 % reduction in acquisition time compared to conventional VIBE. 2) No evidence of a difference was found between conventional VIBE-SPAIR and accelerated VIBE-SPAIRDL regarding lesion detectability or diagnostic confidence.

Evaluating ChatGPT-4V in chest CT diagnostics: a critical image interpretation assessment.

PURPOSE: To assess the diagnostic accuracy of ChatGPT-4V in interpreting a set of four chest CT slices for each case of COVID-19, non-small cell lung cancer (NSCLC), and control cases, thereby evaluating its potential as an AI tool in radiological diagnostics. MATERIALS AND METHODS: In this retrospective study, 60 CT scans from The Cancer Imaging Archive, covering COVID-19, NSCLC, and control cases were analyzed using ChatGPT-4V. A radiologist selected four CT slices from each scan for evaluation. ChatGPT-4V's interpretations were compared against the gold standard diagnoses and assessed by two radiologists. Statistical analyses focused on accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), along with an examination of the impact of pathology location and lobe involvement. RESULTS: ChatGPT-4V showed an overall diagnostic accuracy of 56.76%. For NSCLC, sensitivity was 27.27% and specificity was 60.47%. In COVID-19 detection, sensitivity was 13.64% and specificity of 64.29%. For control cases, the sensitivity was 31.82%, with a specificity of 95.24%. The highest sensitivity (83.33%) was observed in cases involving all lung lobes. The chi-squared statistical analysis indicated significant differences in Sensitivity across categories and in relation to the location and lobar involvement of pathologies. CONCLUSION: ChatGPT-4V demonstrated variable diagnostic performance in chest CT interpretation, with notable proficiency in specific scenarios. This underscores the challenges of cross-modal AI models like ChatGPT-4V in radiology, pointing toward significant areas for improvement to ensure dependability. The study emphasizes the importance of enhancing these models for broader, more reliable medical use.

Fast 5-minute shoulder MRI protocol with accelerated TSE-sequences and deep learning image reconstruction for the assessment of shoulder pain at 1.5 and 3 Tesla.

Purpose: The objective of this study was to implement a 5-minute MRI protocol for the shoulder in routine clinical practice consisting of accelerated 2D turbo spin echo (TSE) sequences with deep learning (DL) reconstruction at 1.5 and 3 Tesla, and to compare the image quality and diagnostic performance to that of a standard 2D TSE protocol. Methods: Patients undergoing shoulder MRI between October 2020 and June 2021 were prospectively enrolled. Each patient underwent two MRI examinations: first a standard, fully sampled TSE (TSES) protocol reconstructed with a standard reconstruction followed by a second fast, prospectively undersampled TSE protocol with a conventional parallel imaging undersampling pattern reconstructed with a DL reconstruction (TSEDL). Image quality and visualization of anatomic structures as well as diagnostic performance with respect to shoulder lesions were assessed using a 5-point Likert-scale (5 = best). Interchangeability analysis, Wilcoxon signed-rank test and kappa statistics were performed to compare the two protocols. Results: A total of 30 participants was included (mean age 50±15 years; 15 men). Overall image quality was evaluated to be superior in TSEDL versus TSES (p<0.001). Noise and edge sharpness were evaluated to be significantly superior in TSEDL versus TSES (noise: p<0.001, edge sharpness: p<0.05). No difference was found concerning qualitative diagnostic confidence, assessability of anatomical structures (p>0.05), and quantitative diagnostic performance for shoulder lesions when comparing the two sequences. Conclusions: A fast 5-minute TSEDL MRI protocol of the shoulder is feasible in routine clinical practice at 1.5 and 3 T, with interchangeable results concerning the diagnostic performance, allowing a reduction in scan time of more than 50% compared to the standard TSES protocol.

Differentiation of Hamartomas and Malignant Lung Tumors in Single-Phased Dual-Energy Computed Tomography.

This study investigated the efficacy of single-phase dual-energy CT (DECT) in differentiating pulmonary hamartomas from malignant lung lesions using virtual non-contrast (VNC), iodine, and fat quantification. Forty-six patients with 47 pulmonary lesions (mean age: 65.2 ± 12.1 years; hamartomas-to-malignant lesions = 22:25; male: 67%) underwent portal venous DECT using histology, PET-CT and follow-up CTs as a reference. Quantitative parameters such as VNC, fat fraction, iodine density and CT mixed values were statistically analyzed. Significant differences were found in fat fractions (hamartomas: 48.9%; malignancies: 22.9%; p ≤ 0.0001) and VNC HU values (hamartomas: -20.5 HU; malignancies: 17.8 HU; p ≤ 0.0001), with hamartomas having higher fat content and lower VNC HU values than malignancies. CT mixed values also differed significantly (p ≤ 0.0001), but iodine density showed no significant differences. ROC analysis favored the fat fraction (AUC = 96.4%; sensitivity: 100%) over the VNC, CT mixed value and iodine density for differentiation. The study concludes that the DECT-based fat fraction is superior to the single-energy CT in differentiating between incidental pulmonary hamartomas and malignant lesions, while post-contrast iodine density is ineffective for differentiation.

Shortening Acquisition Time and Improving Image Quality for Pelvic MRI Using Deep Learning Reconstruction for Diffusion-Weighted Imaging at 1.5 T.

RATIONALE AND OBJECTIVES: To determine the impact on acquisition time reduction and image quality of a deep learning (DL) reconstruction for accelerated diffusion-weighted imaging (DWI) of the pelvis at 1.5 T compared to standard DWI. MATERIALS AND METHODS: A total of 55 patients (mean age, 61 ± 13 years; range, 27-89; 20 men, 35 women) were consecutively included in this retrospective, monocentric study between February and November 2022. Inclusion criteria were (1) standard DWI (DWIS) in clinically indicated magnetic resonance imaging (MRI) at 1.5 T and (2) DL-reconstructed DWI (DWIDL). All patients were examined using the institution's standard MRI protocol according to their diagnosis including DWI with two different b-values (0 and 800 s/mm2) and calculation of apparent diffusion coefficient (ADC) maps. Image quality was qualitatively assessed by four radiologists using a visual 5-point Likert scale (5 = best) for the following criteria: overall image quality, noise level, extent of artifacts, sharpness, and diagnostic confidence. The qualitative scores for DWIS and DWIDL were compared with the Wilcoxon signed-rank test. RESULTS: The overall image quality was evaluated to be significantly superior in DWIDL compared to DWIS for b = 0 s/mm2, b = 800 s/mm2, and ADC maps by all readers (P < .05). The extent of noise was evaluated to be significantly less in DWIDL compared to DWIS for b = 0 s/mm2, b = 800 s/mm2, and ADC maps by all readers (P < .001). No significant differences were found regarding artifacts, lesion detectability, sharpness of organs, and diagnostic confidence (P > .05). Acquisition time for DWIS was 2:06 minutes, and simulated acquisition time for DWIDL was 1:12 minutes. CONCLUSION: DL image reconstruction improves image quality, and simulation results suggest that a reduction in acquisition time for diffusion-weighted MRI of the pelvis at 1.5 T is possible.

Local MRI before and after Tumor Resection in Neuroblastoma: Impact of Residual Disease on Event Free Survival.

(1) Background: The study aimed to investigate the influence of MRI-defined residual disease on local tumor control after resection of neuroblastic tumors in patients without routine adjuvant radiotherapy. (2) Methods: Patients, who underwent tumor resection between 2009 and 2019 and received a pre- and postoperative MRI, were included in this retrospective single-center study. Measurement of residual disease (RD) was performed using standardized criteria. Primary endpoint was the local or combined (local and metastatic) event free survival (EFS). (3) Results: Forty-one patients (20 female) with median age of 39 months were analyzed. Risk group analysis showed eleven low-, eight intermediate-, and twenty-two high-risk patients (LR, IR, HR). RD was found in 16 cases by MRI. A local or combined relapse or progression was found in nine patients of whom eight patients had RD (p = 0.0004). From the six patients with local or combined relapse in the HR group, five had RD (p = 0.005). Only one of 25 patients without RD had a local event. Mean EFS (month) was significantly higher if MRI showed no residual tumor (81 ± 5 vs. 43 ± 9; p = 0.0014) for the total cohort and the HR subgroup (62 ± 7 vs. 31 ± 11; p = 0.016). (4) Conclusions: In our series, evidence of residual tumor, detectable by MRI, was associated with insufficient local control, resulting in relapses or local progression in 50% of patients. Only one of the patients without residual tumor had a local relapse.

Image Quality and Diagnostic Performance of Accelerated 2D Hip MRI with Deep Learning Reconstruction Based on a Deep Iterative Hierarchical Network.

OBJECTIVES: Hip MRI using standard multiplanar sequences requires long scan times. Accelerating MRI is accompanied by reduced image quality. This study aimed to compare standard two-dimensional (2D) turbo spin echo (TSE) sequences with accelerated 2D TSE sequences with deep learning (DL) reconstruction (TSEDL) for routine clinical hip MRI at 1.5 and 3 T in terms of feasibility, image quality, and diagnostic performance. MATERIAL AND METHODS: In this prospective, monocentric study, TSEDL was implemented clinically and evaluated in 14 prospectively enrolled patients undergoing a clinically indicated hip MRI at 1.5 and 3T between October 2020 and May 2021. Each patient underwent two examinations: For the first exam, we used standard sequences with generalized autocalibrating partial parallel acquisition reconstruction (TSES). For the second exam, we implemented prospectively undersampled TSE sequences with DL reconstruction (TSEDL). Two radiologists assessed the TSEDL and TSES regarding image quality, artifacts, noise, edge sharpness, diagnostic confidence, and delineation of anatomical structures using an ordinal five-point Likert scale (1 = non-diagnostic; 2 = poor; 3 = moderate; 4 = good; 5 = excellent). Both sequences were compared regarding the detection of common pathologies of the hip. Comparative analyses were conducted to assess the differences between TSEDL and TSES. RESULTS: Compared with TSES, TSEDL was rated to be significantly superior in terms of image quality (p ≤ 0.020) with significantly reduced noise (p ≤ 0.001) and significantly improved edge sharpness (p = 0.003). No difference was found between TSES and TSEDL concerning the extent of artifacts, diagnostic confidence, or the delineation of anatomical structures (p > 0.05). Example acquisition time reductions for the TSE sequences of 52% at 3 Tesla and 70% at 1.5 Tesla were achieved. CONCLUSION: TSEDL of the hip is clinically feasible, showing excellent image quality and equivalent diagnostic performance compared with TSES, reducing the acquisition time significantly.

Faster Elbow MRI with Deep Learning Reconstruction-Assessment of Image Quality, Diagnostic Confidence, and Anatomy Visualization Compared to Standard Imaging.

OBJECTIVE: The objective of this study was to evaluate a deep learning (DL) reconstruction for turbo spin echo (TSE) sequences of the elbow regarding image quality and visualization of anatomy. MATERIALS AND METHODS: Between October 2020 and June 2021, seventeen participants (eight patients, nine healthy subjects; mean age: 43 ± 16 (20-70) years, eight men) were prospectively included in this study. Each patient underwent two examinations: standard MRI, including TSE sequences reconstructed with a generalized autocalibrating partial parallel acquisition reconstruction (TSESTD), and prospectively undersampled TSE sequences reconstructed with a DL reconstruction (TSEDL). Two radiologists evaluated the images concerning image quality, noise, edge sharpness, artifacts, diagnostic confidence, and delineation of anatomical structures using a 5-point Likert scale, and rated the images concerning the detection of common pathologies. RESULTS: Image quality was significantly improved in TSEDL (mean 4.35, IQR 4-5) compared to TSESTD (mean 3.76, IQR 3-4, p = 0.008). Moreover, TSEDL showed decreased noise (mean 4.29, IQR 3.5-5) compared to TSESTD (mean 3.35, IQR 3-4, p = 0.004). Ratings for delineation of anatomical structures, artifacts, edge sharpness, and diagnostic confidence did not differ significantly between TSEDL and TSESTD (p > 0.05). Inter-reader agreement was substantial to almost perfect (κ = 0.628-0.904). No difference was found concerning the detection of pathologies between the readers and between TSEDL and TSESTD. Using DL, the acquisition time could be reduced by more than 35% compared to TSESTD. CONCLUSION: TSEDL provided improved image quality and decreased noise while receiving equal ratings for edge sharpness, artifacts, delineation of anatomical structures, diagnostic confidence, and detection of pathologies compared to TSESTD. Providing more than a 35% reduction of acquisition time, TSEDL may be clinically relevant for elbow imaging due to increased patient comfort and higher patient throughput.

Combined Metabolic and Functional Tumor Volumes on [18F]FDG-PET/MRI in Neuroblastoma Using Voxel-Wise Analysis.

PURPOSE: The purpose of our study was to evaluate the association between the [18F]FDG standard uptake value (SUV) and the apparent diffusion coefficient (ADC) in neuroblastoma (NB) by voxel-wise analysis. METHODS: From our prospective observational PET/MRI study, a subcohort of patients diagnosed with NB with both baseline imaging and post-chemotherapy imaging was further investigated. After registration and tumor segmentation, metabolic and functional tumor volumes were calculated from the ADC and SUV values using dedicated software allowing for voxel-wise analysis. Under the mean of thresholds, each voxel was assigned to one of three virtual tissue groups: highly vital (v) (low ADC and high SUV), possibly low vital (lv) (high ADC and low SUV), and equivocal (e) with high ADC and high SUV or low ADC and low SUV. Moreover, three clusters were generated from the total tumor volumes using the method of multiple Gaussian distributions. The Pearson's correlation coefficient between the ADC and the SUV was calculated for each group. RESULTS: Out of 43 PET/MRIs in 21 patients with NB, 16 MRIs in 8 patients met the inclusion criteria (PET/MRIs before and after chemotherapy). The proportion of tumor volumes were 26%, 36%, and 38% (v, lv, e) at baseline, 0.03%, 66%, and 34% after treatment in patients with response, and 42%, 25%, and 33% with progressive disease, respectively. In all clusters, the ADC and the SUV correlated negatively. In the cluster that corresponded to highly vital tissue, the ADC and the SUV showed a moderate negative correlation before treatment (R = -0.18; p < 0.0001) and the strongest negative correlation after treatment (R = -0.45; p < 0.0001). Interestingly, only patients with progression (n = 2) under therapy had a relevant part in this cluster post-treatment. CONCLUSION: Our results indicate that voxel-wise analysis of the ADC and the SUV is feasible and can quantify the different quality of tissue in neuroblastic tumors. Monitoring ADCs as well as SUV levels can quantify tumor dynamics during therapy.

Novel deep-learning-based diffusion weighted imaging sequence in 1.5 T breast MRI.

PURPOSE: This study aimed to assess the technical feasibility, the impact on image quality, and the acquisition time (TA) of a new deep-learning-based reconstruction algorithm in diffusion weighted imaging (DWI) of breast magnetic resonance imaging (MRI). METHODS: Retrospective analysis of 55 female patients who underwent breast DWI at 1.5 T. Raw data were reconstructed using a deep-learning (DL) reconstruction algorithm on a subset of the acquired averages, therefore a reduction of TA. Clinically used standard DWI sequence (DWIStd) and the DL-reconstructed images (DWIDL) were compared. Two radiologists rated the image quality of b800 and ADC images, using a Likert-scale from 1 to 5 with 5 being considered perfect image quality. Signal intensities were measured by placing a region of interest (ROI) at the same position in both sequences. RESULTS: TA was reduced by 40 % in DWIDL, compared to DWIStd, DWIDL improved noise and sharpness while maintaining contrast, the level of artifacts, and diagnostic confidence. There were no differences regarding the signal intensity values of the apparent diffusion coefficient (ADC), (p = 0.955), b50-values (p = 0.070) and b800-values (p = 0.415) comparing standard and DL-imaging. Lesion assessment showed no differences regarding the number of lesions in ADC and DWI (both p = 1.000) and regarding the lesion diameter in DWI (p = 0.961;0.972) and ADC (p = 0.961;0.972). CONCLUSIONS: The novel deep-learning-based reconstruction algorithm significantly reduces TA in breast DWI, while improving sharpness, reducing noise, and maintaining a comparable level of image quality, artifacts, contrast, and diagnostic confidence. DWIDL does not influence the quantifiable parameters.

Accelerated diffusion-weighted imaging of the prostate using deep learning image reconstruction: A retrospective comparison with standard diffusion-weighted imaging.

PURPOSE: Routine multiparametric MRI of the prostate reduces overtreatment and increases sensitivity in the diagnosis of the most common solid cancer in men. However, the capacity of MRI systems is limited. Here we investigate the ability of deep learning image reconstruction to accelerate time consuming diffusion-weighted imaging (DWI) acquisition while maintaining diagnostic image quality. METHOD: In this retrospective study, raw data of DWI sequences of consecutive patients undergoing MRI of the prostate at a tertiary care hospital in Germany were reconstructed using standard and deep learning reconstruction. To simulate a shortening of acquisition times by 39 %, one instead of two and six instead of ten averages were used in the reconstruction of b = 0 and 1000 s/mm2 images, respectively. Image quality was assessed by three radiologists and objective image quality metrics. RESULTS: After the application of exclusion criteria, 35 out of 147 patients examined between September 2022 and January 2023 were included in this study. The radiologists perceived less image noise on deep learning reconstructed images at b = 0 s/mm2 images and ADC maps with good inter-reader agreement. Signal-to-noise ratios were similar overall with discretely reduced values in the transitional zone after deep learning reconstruction. CONCLUSIONS: An acquisition time reduction of 39 % without loss in image quality is feasible in DWI of the prostate when using deep learning image reconstruction.

Deep Learning MRI Reconstruction for Accelerating Turbo Spin Echo Hand and Wrist Imaging: A Comparison of Image Quality, Visualization of Anatomy, and Detection of Common Pathologies with Standard Imaging.

RATIONALE AND OBJECTIVES: Magnetic resonance imaging (MRI) of the hand and wrist is a routine MRI examination and takes about 15-20 minutes, which can lead to problems resulting from the relatively long scan time, such as decreased image quality due to motion artifacts and lower patient throughput. The objective of this study was to evaluate a deep learning (DL) reconstruction for turbo spin echo (TSE) sequences of the hand and wrist regarding image quality, visualization of anatomy, and diagnostic performance concerning common pathologies. MATERIALS AND METHODS: Twenty-one patients (mean age: 43 ± 19 [19-85] years, 10 men, 11 female) were prospectively enrolled in this study between October 2020 and June 2021. Each participant underwent two MRI protocols: first, standard fully sampled TSE sequences reconstructed with a standard GRAPPA reconstruction (TSES) and second, prospectively undersampled TSE sequences using a conventional parallel imaging undersampling pattern reconstructed with a DL reconstruction (TSEDL). Both protocols were acquired consecutively in one examination. Two experienced MSK-imaging radiologists qualitatively evaluated the images concerning image quality, noise, edge sharpness, artifacts, and diagnostic confidence, as well as the delineation of anatomical structures (triangular fibrocartilage complex, tendon of the extensor carpi ulnaris muscle, extrinsic and intrinsic ligaments, median nerve, cartilage) using a five-point Likert scale and assessed common pathologies. Wilcoxon signed-rank test and kappa statistics were performed to compare the sequences. RESULTS: Overall image quality, artifacts, delineation of anatomical structures, and diagnostic confidence of TSEDL were rated to be comparable to TSES (p > 0.05). Additionally, TSEDL showed decreased image noise (4.90, median 5, IQR 5-5) compared to TSES (4.52, median 5, IQR 4-5, p < 0.05) and improved edge sharpness (TSEDL: 4.10, median 4, IQR 3.5-5; TSES: 3.57, median 4, IQR 3-4; p < 0.05). Inter- and intrareader agreement was substantial to almost perfect (κ = 0.632-1.000) for the detection of common pathologies. Time of acquisition could be reduced by more than 60% with the protocol using TSEDL. CONCLUSION: Compared to TSES, TSEDL provided decreased noise and increased edge sharpness, equal image quality, delineation of anatomical structures, detection of pathologies, and diagnostic confidence. Therefore, TSEDL may be clinically relevant for hand and wrist imaging, as it reduces examination time by more than 60%, thus increasing patient comfort and patient throughput.

Acquisition time reduction of diffusion-weighted liver imaging using deep learning image reconstruction.

PURPOSE: The purpose of this study was to investigate the impact of deep learning accelerated diffusion-weighted imaging (DWIDL) in 1.5-T liver MRI on image quality, sharpness, and diagnostic confidence. MATERIALS AND METHODS: One-hundred patients who underwent liver MRI at 1.5-T including DWI with two different b-values (50 and 800 s/mm²) between February and April 2022 were retrospectively included. There were 54 men and 46 women, with a mean age of 59 ± 14 (SD) years (range: 21-88 years). The single average raw data were retrospectively processed using a deep learning (DL) image reconstruction algorithm leading to a simulated acquisition time of 1 min 28 s for DWIDL as compared to 2 min 31 s for standard DWI (DWIStd) via reduction of signal averages. All DWI datasets were reviewed by four radiologists using a Likert scale ranging from 1-4 using the following criteria: noise level, extent of artifacts, sharpness, overall image quality, and diagnostic confidence. Furthermore, quantitative assessment of noise and signal-to-noise ratio (SNR) was performed via regions of interest. RESULTS: No significant differences were found regarding artifacts and overall image quality (P > 0.05). Noise measurements for the spleen, liver, and erector spinae muscles revealed significantly lower noise for DWIDL versus DWIStd (P < 0.001). SNR measurements in the above-mentioned tissues also showed significantly superior results for DWIDL versus DWIStd for b = 50 s/mm² and ADC maps (all P < 0.001). For b = 800 s/mm², significantly superior results were found for the spleen, right hemiliver, and erector spinae muscles. CONCLUSIONS: DL image reconstruction of liver DWI at 1.5-T is feasible including significant reduction of acquisition time without compromised image quality.

Thin-Slice Prostate MRI Enabled by Deep Learning Image Reconstruction.

OBJECTIVES: Thin-slice prostate MRI might be beneficial for prostate cancer diagnostics. However, prolongation of acquisition time is a major drawback of thin-slice imaging. Therefore, the purpose of this study was to investigate the impact of a thin-slice deep learning accelerated T2-weighted (w) TSE imaging sequence (T2DLR) of the prostate as compared to conventional T2w TSE imaging (T2S). MATERIALS AND METHODS: Thirty patients were included in this prospective study at one university center after obtaining written informed consent. T2S (3 mm slice thickness) was acquired first in three orthogonal planes followed by thin-slice T2DLR (2 mm slice thickness) in axial plane. Acquisition time of axial conventional T2S was 4:12 min compared to 4:37 min for T2DLR. Imaging datasets were evaluated by two radiologists using a Likert-scale ranging from 1-4, with 4 being the best regarding the following parameters: sharpness, lesion detectability, artifacts, overall image quality, and diagnostic confidence. Furthermore, preference of T2S versus T2DLR was evaluated. RESULTS: The mean patient age was 68 ± 8 years. Sharpness of images and lesion detectability were rated better in T2DLR with a median of 4 versus a median of 3 in T2S (p < 0.001 for both readers). Image noise was evaluated to be significantly worse in T2DLR as compared to T2S (p < 0.001 and p = 0.021, respectively). Overall image quality was also evaluated to be superior in T2DLR versus T2S with a median of 4 versus 3 (p < 0.001 for both readers). Both readers chose T2DLR in 29 cases as their preference. CONCLUSIONS: Thin-slice T2DLR of the prostate provides a significant improvement of image quality without significant prolongation of acquisition time.

Application of deep learning-based super-resolution to T1-weighted postcontrast gradient echo imaging of the chest.

OBJECTIVES: A deep learning-based super-resolution for postcontrast volume-interpolated breath-hold examination (VIBE) of the chest was investigated in this study. Aim was to improve image quality, noise, artifacts and diagnostic confidence without change of acquisition parameters. MATERIALS AND METHODS: Fifty patients who received VIBE postcontrast imaging of the chest at 1.5 T were included in this retrospective study. After acquisition of the standard VIBE (VIBES), a novel deep learning-based algorithm and a denoising algorithm were applied, resulting in enhanced images (VIBEDL). Two radiologists qualitatively evaluated both datasets independently, rating sharpness of soft tissue, vessels, bronchial structures, lymph nodes, artifacts, cardiac motion artifacts, noise levels and overall diagnostic confidence, using a Likert scale ranging from 1 to 4. In the presence of lung lesions, the largest lesion was rated regarding sharpness and diagnostic confidence using the same Likert scale as mentioned above. Additionally, the largest diameter of the lesion was measured. RESULTS: The sharpness of soft tissue, vessels, bronchial structures and lymph nodes as well as the diagnostic confidence, the extent of artifacts, the extent of cardiac motion artifacts and noise levels were rated superior in VIBEDL (all P < 0.001). There was no significant difference in the diameter or the localization of the largest lung lesion in VIBEDL compared to VIBES. Lesion sharpness as well as detectability was rated significantly better by both readers with VIBEDL (both P < 0.001). CONCLUSION: The application of a novel deep learning-based super-resolution approach in T1-weighted VIBE postcontrast imaging resulted in an improvement in image quality, noise levels and diagnostic confidence as well as in a shortened acquisition time.

Combined Deep Learning-based Super-Resolution and Partial Fourier Reconstruction for Gradient Echo Sequences in Abdominal MRI at 3 Tesla: Shortening Breath-Hold Time and Improving Image Sharpness and Lesion Conspicuity.

RATIONALE AND OBJECTIVES: To investigate the impact of a prototypical deep learning-based super-resolution reconstruction algorithm tailored to partial Fourier acquisitions on acquisition time and image quality for abdominal T1-weighted volume-interpolated breath-hold examination (VIBESR) at 3 Tesla. The standard T1-weighted images were used as the reference standard (VIBESD). MATERIALS AND METHODS: Patients with diverse abdominal pathologies, who underwent a clinically indicated contrast-enhanced abdominal VIBE magnetic resonance imaging at 3T between March and June 2021 were retrospectively included. Following the acquisition of the standard VIBESD sequences, additional images for the non-contrast, dynamic contrast-enhanced and post-contrast T1-weighted VIBE acquisition were retrospectively reconstructed using the same raw data and employing a prototypical deep learning-based super-resolution reconstruction algorithm. The algorithm was designed to enhance edge sharpness by avoiding conventional k-space filtering and to perform a partial Fourier reconstruction in the slice phase-encoding direction for a predefined asymmetric sampling ratio. In the retrospective reconstruction, the asymmetric sampling was realized by omitting acquired samples at the end of the acquisition and therefore corresponding to a shorter acquisition. Four radiologists independently analyzed the image datasets (VIBESR and VIBESD) in a blinded manner. Outcome measures were: sharpness of abdominal organs, sharpness of vessels, image contrast, noise, hepatic lesion conspicuity and size, overall image quality and diagnostic confidence. These parameters were statistically compared and interrater reliability was computed using Fleiss' Kappa and intraclass correlation coefficient (ICC). Finally, the rate of detection of hepatic lesions was documented and was statistically compared using the paired Wilcoxon test. RESULTS: A total of 32 patients aged 59 ± 16 years (23 men (72%), 9 women (28%)) were included. For VIBESR, breath-hold time was significantly reduced by approximately 13.6% (VIBESR 11.9 ± 1.2 seconds vs. VIBESD: 13.9 ± 1.4 seconds, p < 0.001). All readers rated sharpness of abdominal organs, sharpness of vessels to be superior in images with VIBESR (p values ranged between p = 0.005 and p < 0.001). Despite reduction of acquisition time, image contrast, noise, overall image quality and diagnostic confidence were not compromised, as there was no evidence of a difference between VIBESR and VIBESD (p > 0.05). The inter-reader agreement was substantial with a Fleiss' Kappa of >0.7 in all contrast phases. A total of 13 hepatic lesions were analyzed. The four readers observed a superior lesion conspicuity in VIBESR than in VIBESD (p values ranged between p = 0.046 and p < 0.001). In terms of lesion size, there was no significant difference between VIBESD and VIBESR for all readers. Finally, there was an excellent inter-reader agreement regarding lesion size (ICC > 0.9). For all readers, no statistically significant difference was observed regarding detection of hepatic lesions between VIBESD and VIBESR. CONCLUSION: The deep learning-based super-resolution reconstruction with partial Fourier in the slice phase-encoding direction enabled a reduction of breath-hold time and improved image sharpness and lesion conspicuity in T1-weighted gradient echo sequences in abdominal magnetic resonance imaging at 3 Tesla. Faster acquisition time without compromising image quality or diagnostic confidence was possible by using this deep learning-based reconstruction technique.

Deep learning-based super-resolution gradient echo imaging of the pancreas: Improvement of image quality and reduction of acquisition time.

PURPOSE: The purpose of this study was to evaluate the impact of a deep learning-based super-resolution technique on T1-weighted gradient-echo acquisitions (volumetric interpolated breath-hold examination; VIBE) on the assessment of pancreatic MRI at 1.5 T compared to standard VIBE imaging (VIBESTD). MATERIALS AND METHODS: This retrospective single-center study was conducted between April 2021 and October 2021. Fifty patients with a total of 50 detectable pancreatic lesion entities were included in this study. There were 27 men and 23 women, with a mean age of 69 ± 13 (standard deviation [SD]) years (age range: 33-89 years). VIBESTD (precontrast, dynamic, postcontrast) was retrospectively processed with a deep learning-based super-resolution algorithm including a more aggressive partial Fourier setting leading to a simulated acquisition time reduction (VIBESR). Image analysis was performed by two radiologists regarding lesion detectability, noise levels, sharpness and contrast of pancreatic edges, as well as regarding diagnostic confidence using a 5-point Likert-scale with 5 being the best. RESULTS: VIBESR was rated better than VIBESTD by both readers regarding lesion detectability (5 [IQR: 5, 5] vs. 5 [IQR: 4, 5], for reader 1; 5 [IQR: 5, 5] vs. 4 [IQR: 4, 5]) for reader 2; both P <0.001), noise levels (5 [IQR: 5, 5] vs. 5 [IQR: 4, 5] for reader 1; 5 [IQR: 5, 5] vs. 4 [IQR: 4, 5] for reader 2; both P <0.001), sharpness and contrast of pancreatic edges (5 [IQR: 5, 5] vs. 5 [IQR: 4, 5] for reader 1; 5 [IQR: 5, 5] vs. 4 [IQR: 4, 5] for reader 2; both P <0.001), as well as regarding diagnostic confidence (5 [IQR: 5, 5] vs. 5 [IQR: 4, 5] for reader 1; 5 [IQR: 5, 5] vs. 4 [IQR: 4, 5] for reader 2; both P <0.001). There were no significant differences between lesion sizes as measured by the two readers on VIBESR and VIBESTD images (P > 0.05). The mean acquisition time for VIBESTD (15 ± 1 [SD] s; range: 11-16 s) was longer than that for VIBESR (13 ± 1 [SD] s; range: 11-14 s) (P < 0.001). CONCLUSION: Our results indicate that the newly developed deep learning-based super-resolution algorithm adapted to partial Fourier acquisitions has a positive influence not only on shortening the examination time but also on improvement of image quality in pancreatic MRI.

Deep Learning Reconstruction for Accelerated Spine MRI: Prospective Analysis of Interchangeability.

Background Deep learning (DL)-based MRI reconstructions can reduce examination times for turbo spin-echo (TSE) acquisitions. Studies that prospectively employ DL-based reconstructions of rapidly acquired, undersampled spine MRI are needed. Purpose To investigate the diagnostic interchangeability of an unrolled DL-reconstructed TSE (hereafter, TSEDL) T1- and T2-weighted acquisition method with standard TSE and to test their impact on acquisition time, image quality, and diagnostic confidence. Materials and Methods This prospective single-center study included participants with various spinal abnormalities who gave written consent from November 2020 to July 2021. Each participant underwent two MRI examinations: standard fully sampled T1- and T2-weighted TSE acquisitions (reference standard) and prospectively undersampled TSEDL acquisitions with threefold and fourfold acceleration. Image evaluation was performed by five readers. Interchangeability analysis and an image quality-based analysis were used to compare the TSE and TSEDL images. Acquisition time and diagnostic confidence were also compared. Interchangeability was tested using the individual equivalence index regarding various degenerative and nondegenerative entities, which were analyzed on each vertebra and defined as discordant clinical judgments of less than 5%. Interreader and intrareader agreement and concordance (κ and Kendall τ and W statistics) were computed and Wilcoxon and McNemar tests were used. Results Overall, 50 participants were evaluated (mean age, 46 years ± 18 [SD]; 26 men). The TSEDL method enabled up to a 70% reduction in total acquisition time (100 seconds for TSEDL vs 328 seconds for TSE, P < .001). All individual equivalence indexes were less than 4%. TSEDL acquisition was rated as having superior image noise by all readers (P < .001). No evidence of a difference was found between standard TSE and TSEDL regarding frequency of major findings, overall image quality, or diagnostic confidence. Conclusion The deep learning (DL)-reconstructed turbo spin-echo (TSE) method was found to be interchangeable with standard TSE for detecting various abnormalities of the spine at MRI. DL-reconstructed TSE acquisition provided excellent image quality, with a 70% reduction in examination time. German Clinical Trials Register no. DRKS00023278 © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Hallinan in this issue.

Comprehensive Clinical Evaluation of a Deep Learning-Accelerated, Single-Breath-Hold Abdominal HASTE at 1.5 T and 3 T.

To evaluate the clinical performance of a deep learning-accelerated single-breath-hold half-Fourier acquisition single-shot turbo spin echo (HASTEDL)-sequence for T2-weighted fat-suppressed MRI of the abdomen at 1.5 T and 3 T in comparison to standard T2-weighted fat-suppressed multi-shot turbo spin echo-sequence. A total of 320 patients who underwent a clinically indicated liver MRI at 1.5 T and 3 T between August 2020 and February 2021 were enrolled in this single-center, retrospective study. HASTEDL and standard sequences were assessed regarding overall and organ-based image quality, noise, contrast, sharpness, artifacts, diagnostic confidence, as well as lesion detectability using a Likert scale ranging from 1 to 4 (4 = best). The number of visible lesions of each organ was counted and the largest diameter of the major lesion was measured. HASTEDL showed excellent image quality (median 4, interquartile range 3-4), although BLADE (median 4, interquartile range 4-4) was rated significantly higher for overall and organ-based image quality of the adrenal gland (P < .001), contrast (P < 0.001), sharpness (P < 0.001), artifacts (P < 0.001), as well as diagnostic confidence (P < .001). No significant differences were found concerning noise (P = 0.886), organ-based image quality of the liver, pancreas, spleen, and kidneys (P = 0.120-0.366), number and measured diameter of the detected lesions (ICC = 0.972-1.0). Reduction of the aquisition time (TA) was at least 89% for 1.5 T images and 86% for 3 T images. HASTEDL provided excellent image quality, good diagnostic confidence and lesion detection compared to a standard T2-sequences, allowing an eminent reduction of the acquisition time.

Background enhancement in contrast-enhanced spectral mammography (CESM): are there qualitative and quantitative differences between imaging systems?

OBJECTIVE: To evaluate the impact of the digital mammography imaging system on overall background enhancement on recombined contrast-enhanced spectral mammography (CESM) images, the overall background enhancement of two different mammography systems was compared. METHODS: In a retrospective single-center study, CESM images of n = 129 female patients who underwent CESM between 2016 and 2019 were analyzed independently by two radiologists. Two mammography machines of different manufacturers were compared qualitatively using a Likert-scale from 1 (minimal) to 4 (marked overall background enhancement) and quantitatively by placing a region of interest and measuring the intensity enhancement. Lesion conspicuity was analyzed using a Likert-scale from 1 (lesion not reliably distinguishable) to 5 (excellent lesion conspicuity). A multivariate regression was performed to test for potential biases on the quantitative results. RESULTS: Significant differences in qualitative background enhancement measurements between machines A and B were observed for both readers (p = 0.003 and p < 0.001). The quantitative evaluation showed significant differences in background enhancement with an average difference of 75.69 (99%-CI [74.37, 77.02]; p < 0.001). Lesion conspicuity was better for machine A for the first and second reader respectively (p = 0.009 and p < 0.001). The factor machine was the only influencing factor (p < 0.001). The factors contrast agent, breast density, age, and menstrual cycle could be excluded as potential biases. CONCLUSION: Mammography machines seem to significantly influence overall background enhancement qualitatively and quantitatively; thus, an impact on diagnostic accuracy appears possible. KEY POINTS: • Overall background enhancement on CESM differs between different vendors qualitatively and quantitatively. • Our retrospective single-center study showed consistent results of the qualitative and quantitative data analysis of overall background enhancement. • Lesion conspicuity is higher in cases of lower background enhancement on CESM.