Liver Magnetic Resonance Imaging Research Articles

Accelerated Diffusion-Weighted Magnetic Resonance Imaging of the Liver at 1.5 T With Deep Learning-Based Image Reconstruction: Impact on Image Quality and Lesion Detection.

To perform image quality comparison between deep learning-based multiband diffusion-weighted sequence (DL-mb-DWI), accelerated multiband diffusion-weighted sequence (accelerated mb-DWI), and conventional multiband diffusion-weighted sequence (conventional mb-DWI) in patients undergoing clinical liver magnetic resonance imaging (MRI). Fifty consecutive patients who underwent clinical MRI of the liver at a 1.5-T scanner, between September 1, 2021, and January 31, 2022, were included in this study. Three radiologists independently reviewed images using a 5-point Likert scale for artifacts and image quality factors, in addition to assessing the presence of liver lesions and lesion conspicuity. DL-mb-DWI acquisition time was 65.0 ± 2.4 seconds, significantly (P < 0.001) shorter than conventional mb-DWI (147.5 ± 19.2 seconds) and accelerated mb-DWI (94.3 ± 1.8 seconds). DL-mb-DWI received significantly higher scores than conventional mb-DWI for conspicuity of the left lobe (P < 0.001), sharpness of intrahepatic vessel margin (P < 0.001), sharpness of the pancreatic contour (P < 0.001), in-plane motion artifact (P = 0.002), and overall image quality (P = 0.005) by reader 2. DL-mb-DWI received significantly higher scores for conspicuity of the left lobe (P = 0.006), sharpness of the pancreatic contour (P = 0.020), and in-plane motion artifact (P = 0.042) by reader 3. DL-mb-DWI received significantly higher scores for strength of fat suppression (P = 0.004) and sharpness of the pancreatic contour (P = 0.038) by reader 1. The remaining quality parameters did not reach statistical significance for reader 1. Novel diffusion-weighted MRI sequence with deep learning-based image reconstruction demonstrated significantly decreased acquisition times compared with conventional and accelerated mb-DWI sequences, while maintaining or improving image quality for routine abdominal MRI. DL-mb-DWI offers a potential alternative to conventional mb-DWI in routine clinical liver MRI.

Using GPT-4 for LI-RADS feature extraction and categorization with multilingual free-text reports.

The Liver Imaging Reporting and Data System (LI-RADS) offers a standardized approach for imaging hepatocellular carcinoma. However, the diverse styles and structures of radiology reports complicate automatic data extraction. Large language models hold the potential for structured data extraction from free-text reports. Our objective was to evaluate the performance of Generative Pre-trained Transformer (GPT)-4 in extracting LI-RADS features and categories from free-text liver magnetic resonance imaging (MRI) reports. Three radiologists generated 160 fictitious free-text liver MRI reports written in Korean and English, simulating real-world practice. Of these, 20 were used for prompt engineering, and 140 formed the internal test cohort. Seventy-two genuine reports, authored by 17 radiologists were collected and de-identified for the external test cohort. LI-RADS features were extracted using GPT-4, with a Python script calculating categories. Accuracies in each test cohort were compared. On the external test, the accuracy for the extraction of major LI-RADS features, which encompass size, nonrim arterial phase hyperenhancement, nonperipheral 'washout', enhancing 'capsule' and threshold growth, ranged from .92 to .99. For the rest of the LI-RADS features, the accuracy ranged from .86 to .97. For the LI-RADS category, the model showed an accuracy of .85 (95% CI: .76, .93). GPT-4 shows promise in extracting LI-RADS features, yet further refinement of its prompting strategy and advancements in its neural network architecture are crucial for reliable use in processing complex real-world MRI reports.

Added Value of Liver MRI in Patients Eligible for Surgical Resection or Ablation of Colorectal Liver Metastases Based on CT: A Systematic Review and Meta-Analysis

Background: Abdominal computed tomography (CT) is the standard imaging modality for detection and staging in patients with colorectal liver metastases (CRLM). Although liver magnetic resonance imaging (MRI) is superior to CT in detecting small lesions, guidelines are ambiguous regarding the added value of an additional liver MRI in the surgical workup of patients with CRLM. Therefore, this systematic review and meta-analysis aimed to evaluate the clinical added value of liver MRI in patients eligible for resection or ablation of CRLM based on CT. Methods: A systematic search was performed in the PubMed, Embase, and Cochrane Library databases through June 23, 2023. Studies investigating the impact of additional MRI on local treatment plan following CT in patients with CRLM were included. Risk of bias was assessed using the QUADAS-2 tool. The pooled weighted proportions for the primary outcome were calculated using random effect meta-analysis. Results: Overall, 11 studies with 1440 patients were included, of whom 468 patients (32.5%) were assessed for change in local treatment plan. Contrast-enhanced liver MRI was used in 10 studies, including gadoxetic acid in 9 studies. Liver MRI with diffusion-weighted imaging was used in 8 studies. Pooling of data found a 24.12% (95% confidence interval, 15.58%–32.65%) change in the local treatment plan based on the added findings of liver MRI following CT. Sensitivity analysis including 5 studies (268 patients) focusing on monophasic portal venous CT followed by gadoxetic acid-enhanced liver MRI with diffusion-weighted imaging showed a change of local treatment plan of 17.88% (95% confidence interval, 5.14%–30.62%). Conclusions: This systematic review and meta-analysis found that liver MRI changed the preinterventional local treatment plan in approximately one-fifth of patients eligible for surgical resection or ablation of CRLM based on CT. These findings suggest a clinically relevant added value of routine liver MRI in the preinterventional workup of CRLM, which should be confirmed by large prospective studies.

Focal fatty sparing areas of the pediatric steatotic liver: pseudolesions on hepatobiliary phase magnetic resonance images.

Focal fatty sparing in liver can be detected as hyperintense pseudolesions on hepatobiliary phase magnetic resonance imaging (MRI). Distinguishing these pseudolesions from liver lesions may make diagnosis challenging. The aim of this study was to evaluate the imaging features of fatty sparing areas on liver MRI in pediatric patients who have been administered gadoxetate disodium. A total of 63 patients between January 2018 and June 2023 underwent gadoxetate disodium-enhanced liver MRI, and 9 (14%) patients with a focal fatty sparing were included in the study. The fat spared areas were evaluated qualitatively and quantitatively including signal intensity measurements and fat fraction calculations. The liver MRI examinations of 9 patients (5 boys, 4 girls; aged 8-18 years, median age: 14.4) using gadoxetate disodium were evaluated. Based on in-phase and opposed-phase sequences, 13 areas of focal fatty sparing were identified. The mean fat fraction of the liver and fat spared areas were 26.2% (range, 15-47) and 9% (range, 2-17), respectively. All fat spared areas were hyperintense in the hepatobiliary phase images. The mean relative enhancement ratios of the liver and fat spared areas were 0.78 (range, 0.35-1.6) and 1.11 (range, 0.45-1.9), respectively. Focal fatty sparing in liver in children was observed as hyperintense on hepatobiliary phase MRI, and it should not be identified as a focal liver lesion.

Congenital diaphragmatic hernia: automatic lung and liver MRI segmentation with nnU-Net, reproducibility of pyradiomics features, and a machine learning application for the classification of liver herniation

Prenatal assessment of lung size and liver position is essential to stratify congenital diaphragmatic hernia (CDH) fetuses in risk categories, guiding counseling, and patient management. Manual segmentation on fetal MRI provides a quantitative estimation of total lung volume and liver herniation. However, it is time-consuming and operator-dependent. In this study, we utilized a publicly available deep learning (DL) segmentation system (nnU-Net) to automatically contour CDH-affected fetal lungs and liver on MRI sections. Concordance between automatic and manual segmentation was assessed by calculating the Jaccard coefficient. Pyradiomics standard features were then extracted from both manually and automatically segmented regions. The reproducibility of features between the two groups was evaluated through the Wilcoxon rank-sum test and intraclass correlation coefficients (ICCs). We finally tested the reliability of the automatic-segmentation approach by building a ML classifier system for the prediction of liver herniation based on support vector machines (SVM) and trained on shape features computed both in the manual and nnU-Net-segmented organs. We compared the area under the classifier receiver operating characteristic curve (AUC) in the two cases. Pyradiomics features calculated in the manual ROIs were partly reproducible by the same features calculated in nnU-Net segmented ROIs and, when used in the ML procedure, to predict liver herniation (both AUC around 0.85).          Conclusion: Our results suggest that automatic MRI segmentation is feasible, with good reproducibility of pyradiomics features, and that a ML system for liver herniation prediction offers good reliability.          Trial registration: https://clinicaltrials.gov/ct2/show/NCT04609163?term=NCT04609163&draw=2&rank=1; Clinical Trial Identification no. NCT04609163.What is Known:• Magnetic resonance imaging (MRI) is crucial for prenatal congenital diaphragmatic hernia (CDH) assessment. It enables the quantification of the total lung volume and the extent of liver herniation, which are essential for stratifying the severity of CDH, guiding counseling, and patient management.• The manual segmentation of MRI scans is a time-consuming process that is heavily reliant upon the skill set of the operator.What is New:• MRI lung and liver automatic segmentation using the deep learning nnU-Net system is feasible, with good Jaccard coefficient values and satisfactory reproducibility of pyradiomics features compared to manual results.• A feasible ML system for predicting liver herniation could improve prenatal assessments and CDH patient management.

Pancreatic iron in pediatric transfusion-dependent beta-thalassemia patients: A longitudinal MRI study.

In pediatric transfusion-dependent thalassemia (TDT) patients, we evaluated the prevalence, pattern, and clinical associations of pancreatic siderosis and the changes in pancreatic iron levels and their association with baseline and changes in total body iron balance. We considered 86 pediatric TDT patients consecutively enrolled in the Extension-Myocardial Iron Overload in Thalassemia Network. Iron overload (IO) was quantified by R2* magnetic resonance imaging (MRI). Sixty-three (73%) patients had pancreatic IO (R2*>38Hz). Global pancreas R2* values were significantly correlated with mean serum ferritin levels, MRI liver iron concentration (LIC) values, and global heart R2* values. Global pancreas R2* values were significantly higher in patients with altered versus normal glucose metabolism. Thirty-one patients also performed the follow-up MRI at 18±3months. Higher pancreatic R2* values were detected at the follow-up, but the difference versus the baseline MRI was not significant. The 20% of patients with baseline pancreatic IO showed no pancreatic IO at the follow-up. The 46% of patients without baseline pancreatic IO developed pancreatic siderosis. The changes in global pancreas R2* between the two MRIs were not correlated with baseline serum ferritin levels, baseline, final, and changes in MRI LIC values, or baseline pancreatic iron levels. In children with TDT, pancreatic siderosis is a frequent finding associated with hepatic siderosis and represents a risk factor for myocardial siderosis and alterations of glucose metabolism. Iron removal from the pancreas is exceptionally challenging and independent from hepatic iron status.

Multi-average high-acceleration modified volumetric interpolated breath-hold examination (VIBE) for free-breathing multiphase contrast-enhanced liver MRI: a comparative study with breath-hold VIBE.

Breath-hold volumetric interpolated breath-hold examination (BH-VIBE) of multiphase contrast-enhanced liver magnetic resonance imaging (MPCE-LMRI) requires good cooperative individuals to comply with multiple breath-holds. To develop a free-breathing modified VIBE (FB-mVIBE) as a substitute of BH-VIBE in MPCE-LMRI. We modified VIBE with a high acceleration factor (2 × 2) and four averages to produce the mVIBE scan. A total of 90 individuals (40 men; mean age = 54.6 ± 10.0 years) who had received MPCE-LMRI as part of a voluntary health check-up for oncology survey were enrolled. Each participant was scanned in four phases (pre-contrast, arterial phase, venous phase, and delay phase), and each phase had two sequential scans. To encounter the timing effect of contrast enhancement, three scan orders were designed: BH-VIBE and FB-mVIBE (group A, n = 30); BH-VIBE and FB-VIBE (group B, n = 30); and FB-mVIBE and BH-VIBE (group C, n = 30). The comparisons included the objective measurements and 25 visual-score by two abdominal radiologists independently. Consistency between raters was observed for all three sequences (intraclass correlation coefficient [ICC] = 0.741-0.829). For rater 1, the mean scores of FB-mVIBE (23.67 ± 1.32) were equal to those of BH-VIBE (23.83 ± 1.98) in groups C and B (P = 0.852). The mean scores of FB-mVIBE (22.07 ± 3.02), but significantly higher than those of FB-VIBE (14.7 ± 3.41) in groups A and B (P <0.001). Similar scores were found for rater 2. The objective measurement of FB-mVIBE were equal to or higher than BH-VIBE and markedly superior to FB-VIBE. FB-mVIBE is a practical alternative to BH-VIBE for individuals who cannot cooperate with multiple breath-holds for MPCE-LMRI.

Signal Intensity Trajectories Clustering for Liver Vasculature Segmentation and Labeling (LiVaS) on Contrast-Enhanced MR Images: A Feasibility Pilot Study.

This study aims to develop a semiautomated pipeline and user interface (LiVaS) for rapid segmentation and labeling of MRI liver vasculature and evaluate its time efficiency and accuracy against manual reference standard. Retrospective feasibility pilot study. Liver MR images from different scanners from 36 patients were included, and 4/36 patients were randomly selected for manual segmentation as referenced standard. The liver was segmented in each contrast phase and masks registered to the pre-contrast segmentation. Voxel-wise signal trajectories were clustered using the k-means algorithm. Voxel clusters that best segment the liver vessels were selected and labeled by three independent radiologists and a research scientist using LiVaS. Segmentation times were compared using a paired-sample t-test on log-transformed data. The agreement was analyzed qualitatively and quantitatively using DSC for hepatic and portal vein segmentations. The mean segmentation time among four readers was significantly shorter than manual (3.6 ± 1.4 vs. 70.0 ± 29.2min; p < 0.001), even when using a higher number of clusters to enhance accuracy. The DSC for portal and hepatic veins reached up to 0.69 and 0.70, respectively. LiVaS segmentations were overall of good quality, with variations in performance related to the presence/severity of liver disease, acquisition timing, and image quality. Our semi-automated pipeline was robust to different MRI vendors in producing segmentation and labeling of liver vasculature in agreement with expert manual annotations, with significantly higher time efficiency. LiVaS could facilitate the creation of large, annotated datasets for training and validation of neural networks for automated MRI liver vascularity segmentation. HIGHLIGHTS: Key Finding: In this pilot feasibility study, our semiautomated pipeline for segmentation of liver vascularity (LiVaS) on MR images produced segmentations with simultaneous labeling of portal and hepatic veins in good agreement with the manual reference standard but at significantly shorter times (mean LiVaS 3.6 ± 1.4 vs. mean manual 70.0 ± 29.2min; p < 0.001). Importance: LiVaS was robust in producing liver MRI vascular segmentations across images from different scanners in agreement with expert manual annotations, with significant ly higher time efficiency, and therefore potential scalability.

Perfusion MR Imaging of Liver: Principles and Clinical Applications.

Perfusion imaging techniques provide quantitative characterization of tissue microvasculature. Perfusion MR of liver is particularly challenging because of dual afferent flow, need for large organ high-resolution coverage, and significant movement with respiration. The most common MR technique used for quantifying liver perfusion is dynamic contrast-enhanced MR imaging. Here, the authors describe the various perfusion MR models of the liver, the basic concepts behind implementing a perfusion acquisition, and clinical results that have been obtained using these models.

From strength to precision: A systematic review exploring the clinical utility of 7-Tesla magnetic resonance imaging in abdominal imaging.

After approval for clinical use in 2017 early investigations of ultra-high-field abdominal magnetic resonance imaging (MRI) have demonstrated the feasibility as well as diagnostic capabilities of liver, kidney, and prostate MRI at 7-Tesla. However, the elevation of the field strength to 7-Tesla not only brought advantages to abdominal MRI but also presented considerable challenges and drawbacks, primarily stemming from heightened artifacts and limitations in Specific Absorption Rate, etc. Furthermore, evidence in the literature is relatively scarce concerning human studies in comparison to phantom/animal studies which necessitates an investigation into the evidence so far in humans and summarizing all relevant evidence. To offer a comprehensive overview of current literature on clinical abdominal 7T MRI that emphasizes current trends, details relevant challenges, and provides a concise set of potential solutions. This systematic review adheres to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. A PubMed search, utilizing Medical Subject Headings terms such as "7-Tesla" and organ-specific terms, was conducted for articles published between January 1, 1985, and July 25, 2023. Eligibility criteria included studies exploring 7T MRI for imaging human abdominal organs, encompassing various study types (in-vivo/ex-vivo, method development, reviews/meta-analyses). Exclusion criteria involved animal studies and those lacking extractable data. Study selection involved initial identification via title/abstract, followed by a full-text review by two researchers, with discrepancies resolved through discussion. Data extraction covered publication details, study design, population, sample size, 7T MRI protocol, image characteristics, endpoints, and conclusions. The systematic review included a total of 21 studies. The distribution of clinical 7T abdominal imaging studies revealed a predominant focus on the prostate (n = 8), followed by the kidney (n = 6) and the hepatobiliary system (n = 5). Studies on these organs, and in the pancreas, demonstrated clear advantages at 7T. However, small bowel studies showed no significant improvements compared to traditional MRI at 1.5T. The majority of studies evaluated originated from Germany (n = 10), followed by the Netherlands (n = 5), the United States (n = 5), Austria (n = 2), the United Kingdom (n = 1), and Italy (n = 1). Further increase of abdominal clinical MRI field strength to 7T demonstrated high imaging potential, yet also limitations mainly due to the inhomogeneous radiofrequency (RF) excitation field relative to lower field strengths. Hence, further optimization of dedicated RF coil elements and pulse sequences are expected to better optimize clinical imaging at high magnetic field strength.

Deep learning to estimate impaired glucose metabolism from Magnetic Resonance Imaging of the liver: An opportunistic population screening approach.

Diabetes is a global health challenge, and many individuals are undiagnosed and not aware of their increased risk of morbidity/mortality although dedicated tests are available, which indicates the need for novel population-wide screening approaches. Here, we developed a deep learning pipeline for opportunistic screening of impaired glucose metabolism using routine magnetic resonance imaging (MRI) of the liver and tested its prognostic value in a general population setting. In this retrospective study a fully automatic deep learning pipeline was developed to quantify liver shape features on routine MR imaging using data from a prospective population study. Subsequently, the association between liver shape features and impaired glucose metabolism was investigated in individuals with prediabetes, type 2 diabetes and healthy controls without prior cardiovascular diseases. K-medoids clustering (3 clusters) with a dissimilarity matrix based on Euclidean distance and ordinal regression was used to assess the association between liver shape features and glycaemic status. The deep learning pipeline showed a high performance for liver shape analysis with a mean Dice score of 97.0±0.01. Out of 339 included individuals (mean age 56.3±9.1 years; males 58.1%), 79 (23.3%) and 46 (13.6%) were classified as having prediabetes and type 2 diabetes, respectively. Individuals in the high risk cluster using all liver shape features (n = 14) had a 2.4 fold increased risk of impaired glucose metabolism after adjustment for cardiometabolic risk factors (age, sex, BMI, total cholesterol, alcohol consumption, hypertension, smoking and hepatic steatosis; OR 2.44 [95% CI 1.12-5.38]; p = 0.03). Based on individual shape features, the strongest association was found between liver volume and impaired glucose metabolism after adjustment for the same risk factors (OR 1.97 [1.38-2.85]; p<0.001). Deep learning can estimate impaired glucose metabolism on routine liver MRI independent of cardiometabolic risk factors and hepatic steatosis.

Automated MRI liver segmentation for anatomical segmentation, liver volumetry, and the extraction of radiomics.

To develop and evaluate a deep convolutional neural network (DCNN) for automated liver segmentation, volumetry, and radiomic feature extraction on contrast-enhanced portal venous phase magnetic resonance imaging (MRI). This retrospective study included hepatocellular carcinoma patients from an institutional database with portal venous MRI. After manual segmentation, the data was randomly split into independent training, validation, and internal testing sets. From a collaborating institution, de-identified scans were used for external testing. The public LiverHccSeg dataset was used for further external validation. A 3D DCNN was trained to automatically segment the liver. Segmentation accuracy was quantified by theDice similarity coefficient (DSC) with respect to manual segmentation. A Mann-Whitney U test was used to compare the internal and external test sets. Agreement of volumetry and radiomic features was assessed using the intraclass correlation coefficient (ICC). In total, 470 patients met the inclusion criteria (63.9±8.2 years; 376 males) and 20 patients were used for external validation (41±12 years; 13 males). DSC segmentation accuracy of the DCNN was similarly high between the internal (0.97±0.01) and external (0.96±0.03) test sets (p=0.28) and demonstrated robust segmentation performance on public testing (0.93±0.03). Agreement of liver volumetry was satisfactory in the internal (ICC, 0.99), external (ICC, 0.97), and public (ICC, 0.85) test sets. Radiomic features demonstrated excellent agreement in the internal (mean ICC, 0.98±0.04), external (mean ICC, 0.94±0.10), and public (mean ICC, 0.91±0.09) datasets. Automated liver segmentation yields robust and generalizable segmentation performance on MRI data and can be used for volumetry and radiomic feature extraction. Liver volumetry, anatomic localization, and extraction of quantitative imaging biomarkers require accurate segmentation, but manual segmentation is time-consuming. A deep convolutional neural network demonstrates fast and accurate segmentation performance on T1-weighted portal venous MRI. • This deep convolutional neural network yields robust and generalizable liver segmentation performance on internal, external, and public testing data. • Automated liver volumetry demonstrated excellent agreement with manual volumetry. • Automated liver segmentations can be used for robust and reproducible radiomic feature extraction.

[Reduction of Motion Artifacts in Liver MRI Using Deep Learning with High-pass Filtering].

To investigate whether deep learning with high-pass filtering can be used to effectively reduce motion artifacts in magnetic resonance (MR) images of the liver. The subjects were 69 patients who underwent liver MR examination at our hospital. Simulated motion artifact images (SMAIs) were created from non-artifact images (NAIs) and used for deep learning. Structural similarity index measure (SSIM) and contrast ratio (CR) were used to verify the effect of reducing motion artifacts in motion artifact reduction image (MARI) output from the obtained deep learning model. In the visual assessment, reduction of motion artifacts and image sharpness were evaluated between motion artifact images (MAIs) and MARIs. The SSIM values were 0.882 on the MARIs and 0.869 on the SMAIs. There was no statistically significant difference in CR between NAIs and MARIs. The visual assessment showed that MARIs had reduced motion artifacts and improved sharpness compared to MAIs. The learning model in this study is indicated to be reduced motion artifacts without decreasing the sharpness of liver MR images.

Breath-hold High-resolution T1-weighted Gradient Echo Liver MR Imaging with Compressed Sensing Obtained during the Gadoxetic Acid-enhanced Hepatobiliary Phase: Image Quality and Lesion Visibility Compared with a Standard T1-weighted Sequence.

To evaluate the feasibility of breath-hold (BH) high-resolution (HR) T1-weighted gradient echo hepatobiliary phase (HBP) imaging using compressed sensing (CS) in gadoxetic acid-enhanced liver MRI in comparison with standard HBP imaging using parallel imaging (PI). The study included 122 patients with liver tumors with hypointensity in the HBP who underwent both HR HBP imaging with CS and standard HBP imaging with PI. Two radiologists evaluated the liver edge sharpness, hepatic vessel conspicuity, bile duct conspicuity, image noise, and overall image quality, as well as the lesion conspicuity on HR and standard HBP imaging and the contrast-enhanced (CE) MR cholangiography (MRC) image quality reconstructed from HBP images. As a quantitative analysis, the SNR of the liver and the liver to lesion signal intensity ratio (LLSIR) were also determined. The liver edge sharpness, hepatic vessel conspicuity, bile duct conspicuity, and overall image quality as well as the lesion conspicuity and the LLSIR on HR HBP imaging with CS were significantly higher than those on standard HBP imaging (all of P < 0.001). The image quality of CE-MRC reconstructed from HR HBP imaging with CS was also significantly higher than that from standard HBP imaging (P < 0.001). Conversely, the SNR of liver in standard HBP was significantly higher than that in HR HBP with CS (P < 0.001). BH HR HBP imaging with CS provided an improved overall image quality, lesion conspicuity, and CE-MRC visualization when compared with standard HBP imaging without extending the acquisition time.

Application of a deep learning algorithm for three-dimensional T1-weighted gradient-echo imaging of gadoxetic acid-enhanced MRI in patients at a high risk of hepatocellular carcinoma.

To evaluate the efficacy of a vendor-specific deep learning reconstruction algorithm (DLRA) in enhancing image quality and focal lesion detection using three-dimensional T1-weighted gradient-echo images in gadoxetic acid-enhanced liver magnetic resonance imaging (MRI) in patients at a high risk of hepatocellular carcinoma. In this retrospective analysis, 83 high-risk patients with hepatocellular carcinoma underwent gadoxetic acid-enhanced liver MRI using a 3-T scanner. Triple arterial phase, high-resolution portal venous phase, and high-resolution hepatobiliary phase images were reconstructed using conventional reconstruction techniques and DLRA (AIRTM Recon DL; GE Healthcare) for subsequent comparison. Image quality and solid focal lesion detection were assessed by three abdominal radiologists and compared between conventional and DL methods. Focal liver lesion detection was evaluated using figures of merit (FOMs) from a jackknife alternative free-response receiver operating characteristic analysis on a per-lesion basis. DLRA-reconstructed images exhibited significantly improved overall image quality, image contrast, lesion conspicuity, vessel conspicuity, and liver edge sharpness and reduced subjective image noise, ringing artifacts, and motion artifacts compared to conventionally reconstructed images (all P < 0.05). Although there was no significant difference in the FOMs of non-cystic focal liver lesions between the conventional and DL methods, DLRA-reconstructed images showed notably higher pooled sensitivity than conventionally reconstructed images (P < 0.05) in all phases and higher detection rates for viable post-treatment HCCs in the arterial and hepatobiliary phases (all P < 0.05). Implementing DLRA can enhance the image quality in 3D T1-weighted gradient-echo sequences of gadoxetic acid-enhanced liver MRI examinations, leading to improved detection of viable post-treatment HCCs.

Assessment of Risk Factors for Developing Iron Overload in Patients with Pediatric Cancer and Bone Marrow Transplant Survivors: A Single Center Retrospective Chart Review

Background: Pediatric cancer survivors and those with chronic hematologic disorders are at risk for iron overload secondary to frequent transfusion therapy. As this patient population increases in number due to improvement of therapies, it is essential to identify risk factors associated with the development of iron overload. Classifying these risk factors will help improve screening methods and assist in decreasing side effects from iron overload, such as iron deposition resulting in organ damage. Objective: Assessment of risk factors associated with the development of iron overload in pediatric cancer survivors and those who underwent bone marrow transplant. Methods: Retrospective chart review of patients who completed therapy for pediatric malignancy or underwent bone marrow transplant and were screened for iron overload utilizing magnetic resonance imaging (MRI). Iron overload was defined as T2* liver MRI with iron levels of &amp;gt;5mg/g dry weight. Individuals determined to screen positive for iron overload were evaluated for additional demographic and treatment variables, such as race, age at diagnosis, and total transfused blood volume, to assess for risk factors associated with the development of iron overload. Statistical analyses were performed via univariate and multivariate regression models of the variables collected. Results: We report a large cohort of patients with a mean age of 10.6 years who underwent screening T2* liver MRIs. Of 71 patients who underwent liver MRI screening, 89% met criteria for moderately severe iron overload. Blood volume administered (ml) and Non-White Race were found to be statistically significant in association with higher iron content. Mean iron content was 12.9 mg/g in the Non-White group versus 8.8 mg/g in the White group (p=0.0185). For every 100ml/kg increase in the cumulative blood volume transfused, the predicted iron content increased approximately 1 mg/g (p=0.0227); mean transfused blood volume was 340ml/kg among those who met criteria for iron overload versus 224ml/kg among those who did not (p=0.1858). We did not find any correlation between iron content and age at diagnosis, primary diagnosis, BMI at diagnosis, or gender. Conclusions: Iron overload continues to be a risk for pediatric cancer survivors and those who receive chronic transfusions. Our study revealed a statistically significant correlation between increased amount of blood transfusions and development of iron overload; however, no specific quantity of cumulative transfused blood volume was determined to be the threshold for risk. Our study also revealed a statistically significant correlation between Non-White race and the development of iron overload independent of cumulative transfused blood volume. It is possible there are race-based differences in iron metabolism that result in this increased risk. Further studies regarding specific quantities of blood transfusions, assessing physiologic characteristics of the Non-White patient population, and development of iron overload are needed. These findings indicate the importance of continuing to research the development of iron overload in the pediatric cancer and bone marrow transplant population, as well as provide guidance for understanding the risk each unique patient has for developing this complication.

Improvement in Iron Burden in Patients with Transfusion-Dependent β-Thalassemia (TDT) Treated with Betibeglogene Autotemcel (Beti-cel) Gene Therapy: Up to 9 Years of Follow-up

Introduction: Beti-cel is an approved, one-time gene therapy treatment for adult and pediatric patients with TDT. Beti-cel consists of autologous hematopoietic stem and progenitor cells transduced with the BB305 lentiviral vector encoding a modified β-globin gene (β A-T87Q), which produces functional hemoglobin (Hb) and addresses the underlying genetic cause of TDT. While transfusion independence (TI) remains a critical component for assessing patient response to gene therapy, other measures, including markers of iron overload, are important for evaluating the impact of therapy and management of disease burden over time. Previously, we demonstrated that iron markers stabilized in patients who achieved TI and stopped chelation therapy. Here, we report iron management outcomes in patients who completed either a phase 1/2 or phase 3 beti-cel parent study and subsequently enrolled in the long-term follow-up study. Methods: Patients with TDT who completed a parent beti-cel study (phase 1/2 studies: HGB-204 [NCT01745120]; HGB-205 [NCT02151526]; phase 3 studies: HGB-207 [NCT02906202]; HGB-212 [NCT03207009]) could enroll in a long-term study (LTF-303 [NCT02633943]) for up to an additional 13 years of follow-up. In these studies, TI was defined as a weighted average Hb ≥9 g/dL without packed red blood cell transfusions for ≥12 months. Iron removal therapy was at the discretion of the treating physician. Liver and cardiac magnetic resonance imaging evaluated liver iron concentration (LIC) and cardiac T2*; serum ferritin and transferrin receptor (TfR) were assessed at regular intervals according to protocol. Results:As of January 30, 2023, 63 patients received beti-cel; median (range) follow-up was 60.1 (23.8-109.5) months. In phase 3 studies, 37/41 (90%) patients achieved TI and maintained TI through the last follow-up, which was 5+ years for 10 patients ( Figure 1). In phase 1/2 studies, 15/22 (68%) patients achieved TI; 14 of these patients maintained TI through the last follow-up, which was up to 9 years. One patient no longer meets protocol-defined TI as a result of Hb level &amp;lt;9g/dL at year 6 due to acute health events unrelated to β-thalassemia, which were not attributed to loss of beti-cel treatment effect. In total, 51 patients across phase 1/2 and 3 studies achieved and maintained TI through last follow-up with 35/51 (69%) patients off chelation. Thirty of 51 patients had an LIC measurement at month 48. Of these 30 patients, 21 had an LIC &amp;lt;5 mg/g at month 48 post infusion. Among these 21 patients, 5 never restarted chelation after infusion (2 received phlebotomy only), 11 restarted chelation and then were able to discontinue after iron stores normalized (1 also received phlebotomy), and 5 continue to receive chelation (none had phlebotomy). The other 9 patients who had an LIC measurement at month 48 had an iron burden ≥5 mg/g. All 9 of these patients restarted chelation; 4 were able to discontinue (2 had phlebotomy) and 5 continue to receive chelation (1 had phlebotomy). Analysis of iron marker stabilization after discontinuation of chelation will be presented. The median (range) change from baseline in LIC levels at month 48 was -3.9 (-22.3 to 10.5) mg Fe/g dry weight; the median (range) change from baseline in serum ferritin was -2119.0 (-7211 to 3889) pmoL/L. Reductions from baseline in serum TfR were observed at month 36; cardiac T2* remained stable at month 48 ( Table 1). Safety of beti-cel treatment largely reflected the known side effects of hematopoietic stem cell collection and the busulfan conditioning regimen. Conclusion: In this analysis with up to 9 years of follow up, patients treated with beti-cel who achieved TI also demonstrated sustained improvements in iron burden, and the majority of patients were able to stop chelation. Collectively, these results demonstrate the long-term durability and stability of response after beti-cel gene therapy in patients with TDT.

Suboptimal hepatobiliary phase image in gadoxetic acid-enhanced liver MRI for the evaluation of the HCC: Predictive factors.

To determine the relevant laboratory values for hepatobiliary phase (HBP) imaging and predictive factors for suboptimal HBP images on gadoxetic acid-enhanced liver magnetic resonance imaging (MRI) for the evaluation of hepatocellular carcinoma (HCC) in patients with chronic liver disease (CLD). This study included 307 patients with CLD who underwent gadoxetic acid-enhanced liver MRI for HCC evaluation. The liver-portal vein contrast ratio and liver-spleen contrast ratio were calculated from the measurements of the HBP images. In this study, a suboptimal HBP image was defined as the presence of a bright portal vein or a liver-spleen contrast ratio of <1.5. Correlation, comparison, and receiver operating characteristic analyses were performed between the measured parameters on the HBP images and hepatic and renal function tests. The estimated glomerular filtration rate did not correlate with any measured or calculated values on the HBP images. On receiver operating characteristic analysis, the optimal cutoff value for the bright portal vein was an albumin level of 4.05 g/dL (area under the curve, 0.971; sensitivity, 65%; specificity, 82%). The optimal cutoff value of the suboptimal HBP image was a serum direct bilirubin level of 0.83 mg/dL (area under the curve, 0.830; sensitivity, 69%; specificity, 84%). On gadoxetic acid-enhanced MRI for the evaluation of HCC in patients with CLD, suboptimal HBP images were most strongly correlated with serum direct bilirubin levels. Renal function was not associated with suboptimal HBP imaging. Although the sensitivity is low, suboptimal HBP images can be predicted before gadoxetic acid-enhanced liver MRI can be performed.

Prognostic value of liver fibrotic markers in patients with heart failure

Abstract Background One of the major comorbidities in patients with heart failure (HF) is hepatic dysfunction, which may occur with prolonged hepatic hypoperfusion and congestion due to HF. Although several liver fibrotic markers are associated with prognosis in patients with HF, the optimal markers for outcome prediction remain unclear. Purpose The aim of this study was to simultaneously investigate the prognostic value of liver fibrotic markers and the associations between these markers and clinical parameters in HF patients without organic liver disease. Methods We prospectively examined 211 consecutive patients with chronic HF (116 male, mean age 66 ± 14 years, LVEF 48 ± 17%) between April 2018 and August 2021, excluding those with organic liver disease, using liver magnetic resonance imaging and ultrasound. We measured seven representative liver fibrotic markers, including hyaluronic acid, type III procollagen N-terminal peptide (P-III-P), type IV collagen 7s domain, mac-2 binding protein glycosylation isomer (M2BPGi), platelet ratio index (APRI), fibrosis 4 index (FIB-4), and FibroIndex. The primary outcome of interest was the composite of all-cause death and hospitalisation for HF. Results During a median follow-up period of 747 (interquartile range 465-1042) days, the primary outcome occurred in 45 patients (21%), including 15 (7%) all-cause deaths and 30 (14%) hospitalisations for HF. Patients with higher hyaluronic acid and P-III-P levels showed a significantly higher incidence of the primary outcome than those without (P &amp;lt;0.001 and P = 0.005, respectively). Multivariable Cox regression analysis showed that hyaluronic acid and P-III-P levels were independently associated with the risk of the primary outcome (hazard ratio [HR] 1.84, 95% confidence interval [CI] 1.18-2.87 and HR 2.89, 95% CI 1.32-6.34, respectively) even after adjustment for the Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC) risk score, whereas other five liver fibrotic markers were not associated with the primary outcome. Discriminative value of the MAGGIC score for the primary outcome was significantly increased by adding hyaluronic acid and P-III-P levels (Harrel’s c-statistics; from 0.717 to 0.752 and 0.745, respectively) with significant net reclassification improvement (38% and 52%, respectively). Hyaluronic acid was predominantly correlated with age, haemoglobin, serum albumin, troponin T, N-terminal pro-brain natriuretic peptide (NT-proBNP), and pulmonary artery compliance (Figure). P-III-P levels were correlated with haemoglobin, estimated glomerular filtration rate, alkaline phosphatase, and NT-proBNP levels (Figure). Conclusion Among the representative liver fibrotic markers, hyaluronic acid and P-III-P might be optimal for outcome prediction in HF patients without organic liver disease. Hyaluronic acid and P-III-P levels were predominantly associated with several key prognostic indicators of HF simultaneously.

Liver fibrosis MR images classification based on higher-order interaction and sample distribution rebalancing.

The fractal features of liver fibrosis MR images exhibit an irregular fragmented distribution, and the diffuse feature distribution lacks interconnectivity, result- ing in incomplete feature learning and poor recognition accuracy. In this paper, we insert recursive gated convolution into the ResNet18 network to introduce spatial information interactions during the feature learning process and extend it to higher orders using recursion. Higher-order spatial information interactions enhance the correlation between features and enable the neural network to focus more on the pixel-level dependencies, enabling a global interpretation of liver MR images. Additionally, the existence of light scattering and quantum noise during the imaging process, coupled with environmental factors such as breathing artifacts caused by long time breath holding, affects the quality of the MR images. To improve the classification performance of the neural network and better cap- ture sample features, we introduce the Adaptive Rebalance loss function and incorporate the feature paradigm as a learnable adaptive attribute into the angular margin auxiliary function. Adaptive Rebalance loss function can expand the inter-class distance and narrow the intra-class difference to further enhance discriminative ability of the model. We conduct extensive experiments on liver fibrosis MR imaging involving 209 patients. The results demonstrate an average improvement of two percent in recognition accuracy compared to ResNet18. The github is at https://github.com/XZN1233/paper.git.