BACKGROUND: Epicardial adipose tissue volume (EATV) is increasingly recognized as a cardiometabolic risk marker associated with adverse outcomes. The most established approach for EATV quantification is cardiac computed...BACKGROUND: Epicardial adipose tissue volume (EATV) is increasingly recognized as a cardiometabolic risk marker associated with adverse outcomes. The most established approach for EATV quantification is cardiac computed tomography (CT). MRI offers a radiation-free alternative allowing simultaneous assessment of myocardial function and tissue characteristics; however, standardization and validation are limited. PURPOSE: To evaluate a standardized MRI-based method for EATV quantification and determine its agreement with CT. STUDY TYPE: Retrospective. POPULATION: 127 patients with aortic stenosis (AS) (78 ± 6 years; 38% female) who underwent paired CT and MRI and 11 volunteers (74 ± 7 years, 45% female) who underwent repeat MRI after ≥ 6 weeks. FIELD STRENGTH/SEQUENCE: Short-axis balanced steady-state free precession cine sequence at 3T (AS patients) and 1.5T (volunteers). ASSESSMENT: On CT, EATV was quantified by manual delineation of the visceral pericardium and voxel-thresholding (-190 to -30 HU). MRI-based EATV quantification used manual volumetry with delineation of the visceral pericardium and epicardium on end-diastolic short-axis cine stacks. STATISTICAL TESTS: Inter-modality agreement was assessed by Spearman correlation and Bland-Altman analysis. Reproducibility was evaluated in 20 patients using intraclass correlation coefficient (ICC) and coefficient of variation (CoV). Scan-rescan reproducibility for MRI-derived EATV quantification was assessed using ICC and linear regression. p < 0.05 was considered significant. RESULTS: Median EATV was significantly higher on MRI than CT (47 vs. 38 mL/m), with a significant moderate correlation (ρ = 0.627) between measures. Inter- and intra-observer analyses showed excellent reproducibility for both modalities (CT: intra-observer ICC: 0.983, inter-observer ICC: 0.994; MRI: intra-observer ICC: 0.955, inter-observer ICC: 0.970). MRI-derived EATV quantification also showed excellent scan-rescan reproducibility (ICC: 0.985). DATA CONCLUSION: The standardized MRI-based approach enabled highly reproducible EATV measurements with excellent repeatability. Agreement with CT was moderate, with systematically higher values on MRI, limiting direct comparability. EVIDENCE LEVEL: 3. STAGE OF TECHNICAL EFFICACY: 2.
BACKGROUND: Differentiating true progression (TP) from pseudoprogression (PsP) after chemoradiotherapy in gliomas remains challenging because conventional MRI findings overlap. PURPOSE: To assess whether voxel-level radi...BACKGROUND: Differentiating true progression (TP) from pseudoprogression (PsP) after chemoradiotherapy in gliomas remains challenging because conventional MRI findings overlap. PURPOSE: To assess whether voxel-level radiomics habitat metrics improve TP/PsP classification and complement clinical and molecular features. STUDY TYPE: Retrospective, multi-center. SUBJECTS: 193 glioma patients after treatment: 121 in the training set (54.2 ± 11.3 years; 74 men; 85 TP/36 PsP) and 72 in the external testing set (51.4 ± 10.4 years; 44 men; 52 TP/20 PsP). FIELD STRENGTH/SEQUENCE: 3 and 1.5 T; contrast-enhanced T1-weighted spin-echo imaging (T1CE), T2-weighted fast/turbo spin-echo imaging (T2WI), T2-weighted fluid-attenuated inversion recovery fast/turbo spin-echo imaging (T2-FLAIR), diffusion-weighted echo-planar imaging (DWI), and arterial spin labeling (ASL). ASSESSMENT: Voxel-wise radiomics features were extracted from contrast-enhancing tumor ROIs. Gaussian mixture models generated soft habitats, from which voxel-level metrics were calculated. Three models were constructed. The Clinical Model included tumor grade, isocitrate dehydrogenase (IDH) status, O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status, and time interval. Feature selection used least absolute shrinkage and selection operator (LASSO); classifier optimization employed Optuna-based Bayesian methods. STATISTICAL TESTS: Receiver operating characteristic (ROC) curve analysis, calibration curves, decision curve analysis (DCA), and SHAP were used. p < 0.05 indicated significance. RESULTS: In the testing set, the Clinical, Voxel-wise Habitat, and Combined Models achieved areas under the curve (AUCs) of 0.701 (95% confidence interval [CI]: 0.597-0.806), 0.832 (95% CI: 0.736-0.917), and 0.890 (95% CI: 0.819-0.958), respectively. The Combined Model significantly outperformed the clinical model (difference, 0.189). Voxel-wise Habitat versus Clinical (difference, 0.131; p = 0.164) and Combined versus Voxel-wise Habitat comparisons (difference, 0.058; p = 0.105) were not significant. SHAP ranked CBF_habitat_edge_standard_deviation, CBF_habitat_entropy_mean, and T1CE_habitat_edge_standard_deviation as leading contributors. DATA CONCLUSION: Voxel-wise habitat analysis, combined with clinical and molecular features, improved TP/PsP discrimination with interpretable heterogeneity metrics. EVIDENCE LEVEL: 3. TECHNICAL EFFICACY: Stage 3.
BACKGROUND: The T2-FLAIR mismatch (T2FM) sign is a highly specific imaging biomarker for isocitrate dehydrogenase (IDH)-mutant astrocytomas; however, its strict definition limits clinical applicability. PURPOSE: To deter...BACKGROUND: The T2-FLAIR mismatch (T2FM) sign is a highly specific imaging biomarker for isocitrate dehydrogenase (IDH)-mutant astrocytomas; however, its strict definition limits clinical applicability. PURPOSE: To determine whether an expanded T2FM (eT2FM) phenotype could improve identification of IDH-mutant astrocytomas and capture prognostic and biological information. STUDY TYPE: Retrospective. POPULATION: 349 patients (50.14% male; mean age, 41.08 ± 10.40 years) with IDH-mutant astrocytomas. FIELD STRENGTH/SEQUENCE: 3 T, T1-weighted gradient-echo- or spin-echo-based images, and T2-weighted, T2 fluid-attenuated inversion recovery (FLAIR), and contrast-enhanced T1-weighted images mainly acquired using spin-echo-based techniques. ASSESSMENT: The eT2FM phenotype was defined by incorporating spatially heterogeneous T2 FLAIR signals beyond the classic T2FM sign. Survival differences were compared between eT2FM and non-eT2FM tumors. Survival associations were evaluated. Metabolomic profiling was explored. STATISTICAL TESTS: Kaplan-Meier analysis, weighted log-rank test, Cox regression, and metabolomic analyses. Significance was set at p < 0.05. RESULTS: The eT2FM phenotype was identified in 116 of 349 patients (33.24%; mean age, 39.08 ± 9.51 years), exceeding classic T2FM (50/349, 14.33%; mean age, 36.40 ± 9.10 years). Kaplan-Meier analysis showed more favorable overall survival for eT2FM than non-eT2FM tumors, with restricted mean survival time (RMST) up to 36 months of 35.18 and 32.08 months, respectively (absolute difference, 3.10 months). Similar findings were observed in the external validation cohort, in which eT2FM phenotype was identified in 48 of 146 patients (32.88%; mean age, 37.77 ± 11.05 years). Kaplan-Meier analysis also showed more favorable overall survival for eT2FM than non-eT2FM tumors, with RMSTs of 36.00 and 33.81 months, respectively (absolute difference, 2.19 months). The eT2FM phenotype was associated with better prognosis in univariable Cox analysis (HR, 0.265; 95% CI, 0.120-0.585), but not independently in multivariable analysis (HR, 2.014; 95% CI, 0.380-10.673; p = 0.41). Exploratory metabolomics identified 1078 differentially abundant features between groups. DATA CONCLUSION: The eT2FM phenotype extends the clinical utility of classic T2FM sign and delineates a subgroup of IDH-mutant astrocytomas with favorable clinicopathologic features. TECHNICAL EFFICACY: Stage 4.
BACKGROUND: Evaluating pulmonary congestion after exercise in heart failure (HF) may reveal perturbations not apparent at rest and provide insight into dyspnea and exercise intolerance. PURPOSE: To investigate the reprod...BACKGROUND: Evaluating pulmonary congestion after exercise in heart failure (HF) may reveal perturbations not apparent at rest and provide insight into dyspnea and exercise intolerance. PURPOSE: To investigate the reproducibility and exercise-induced alterations in lung water density (LWD) using 3D ultrashort echo time (UTE) MRI. STUDY TYPE: Prospective. FIELD STRENGTH/SEQUENCE: 3-T MRI with a prototype 3D stack-of-spirals UTE Volumetric Interpolated Breath-hold Examination (VIBE) sequence. POPULATION: Sixty-nine HF patients (59 ± 12 years; 39 males) across a range of left ventricular ejection fraction (23 HFpEF, 16 HFmrEF, 18 HFimpEF, and 12 HFrEF); 34 healthy subjects. ASSESSMENT: UTE MRI was performed at rest and after exercise. Images were normalized to yield relative LWD (%). LWD was quantified as the mean normalized pixel intensity within the segmented parenchyma, and lung volume was also measured. Reproducibility was evaluated in a separate scan/rescan study. STATISTICAL TESTS: Shapiro-Wilk test; one-way ANOVA or Kruskal-Wallis test with FDR correction; Student's t-test or Wilcoxon signed-rank test; χ test or Fisher's exact test; Bland-Altman analysis. p < 0.05 was considered significant. RESULTS: In healthy controls, LWD decreased significantly after exercise with an absolute change of -1.25% (interquartile range [IQR]: -2.92% to 0.38%) and relative change of -4.35% (IQR: -9.55% to 1.30%). Compared with controls, HFpEF showed a significantly greater absolute change (0.47%, IQR: -0.91% to 2.10%) and relative change in LWD (1.46%, IQR: -2.65% to 6.45%), not apparent in other HF subgroups (p = 0.053-0.821). Changes in lung volume did not differ between each HF subgroup and healthy controls (p = 0.056-0.498). Bland-Altman analysis demonstrated good agreement for LWD and lung volume. DATA CONCLUSION: Single-breath-hold 3D UTE MRI provides reproducible LWD measurements at rest and enables detection of exercise-induced LWD alterations. This approach captures pulmonary fluid changes following exercise and identifies greater LWD response in HFpEF. TECHNICAL EFFICACY: Stage 1.
BACKGROUND: Thalamic volume loss is one of the most consistent neuroimaging findings in type 1 diabetes mellitus (T1DM). However, the thalamus comprises multiple nuclei with distinct functions, and whether specific nucle...BACKGROUND: Thalamic volume loss is one of the most consistent neuroimaging findings in type 1 diabetes mellitus (T1DM). However, the thalamus comprises multiple nuclei with distinct functions, and whether specific nuclei show differential vulnerability remains unclear. PURPOSE: To investigate the characteristics of intra-thalamic nuclear atrophy in T1DM and evaluate its association with disease duration. FIELD STRENGTH/SEQUENCE: 3.0 T/T1-weighted magnetization-prepared-rapid-acquisition-of-gradient-echo (MPRAGE). STUDY TYPE: Prospective. POPULATION: Forty-two subjects with T1DM (18-68 years, 17 males/25 females) and 39 healthy controls (18-69 years, 16 males/23 females). ASSESSMENT: Multi-atlas-based segmentation was used to quantify volumes of 20 thalamic nuclei. Additionally, volumes of anatomical and functional subregions were derived based on topographic and functional nuclear classifications. STATISTICAL TESTS: Group differences were assessed using general linear models. Associations between thalamic volume loss and disease duration were evaluated via partial correlations and exponential regression models to estimate duration-related atrophy rates. p value < 0.05 was considered significant. RESULTS: Participants with T1DM showed significant bilateral reductions in total thalamic volume (left: 7.6%, Cohen's d = 0.91; right: 8.0%, Cohen's d = 0.95). The most pronounced atrophy was observed in the anterior, medial, and lateral thalamic subregions, with volume reductions of 15.8%, 8.6%, and 7.3%, respectively, particularly in cognition- and motor-related nuclei (9.4% and 9.2%, respectively). Longer disease duration was significantly associated with smaller thalamic volumes in the bilateral whole thalamus (r = -0.48 and r = -0.46), as well as in the Pul (r = -0.64 and r = -0.57), VLp (r = -0.39 and r = -0.38), and MD-Pf nuclei (r = -0.55 and r = -0.37), and in the left LGN (r = -0.43). Estimated disease-duration-related annualized atrophy rates exceeded age-related rates in several left-sided nuclei, including the MD-Pf (-0.67%), Pul (-0.69%), and LGN (-0.75%). DATA CONCLUSION: T1DM is associated with spatially heterogeneous atrophy of thalamic nuclei, with preferential involvement of the anterior, medial, and lateral subregions. Disease duration seems to accelerate neurodegenerative changes. EVIDENCE LEVEL: 2. TECHNICAL EFFICACY: Stage 2.
BACKGROUND: Despite promising results of artificial intelligence (AI) in prostate cancer (PCa) detection, its impact on biparametric MRI (bpMRI) interpretation remains uncertain, especially for readers with limited exper...BACKGROUND: Despite promising results of artificial intelligence (AI) in prostate cancer (PCa) detection, its impact on biparametric MRI (bpMRI) interpretation remains uncertain, especially for readers with limited experience. PURPOSE: To evaluate the effect of AI software assistance on prostate bpMRI interpretation by readers with different levels of prostate MRI experience. STUDY TYPE: Retrospective. POPULATION: Six hundred and forty-six male patients, including 297 with PCa. FIELD STRENGTH/SEQUENCE: 3.0 T; T2-weighted imaging using fast spin echo sequence, diffusion-weighted imaging using single-shot echo-planar imaging. ASSESSMENT: Two experienced readers (8 and 10 years of prostate MRI experience) and two novice-level readers (2 years of general radiology experience; 20-50 prior prostate MRI cases) assessed all examinations twice, without and with AI software (uAI, United Imaging) assistance, in counterbalanced orders with a 4-week washout interval. Lesions were scored using Prostate Imaging Reporting and Data System (PI-RADS) v2.1 at ≥ 3 and ≥ 4 thresholds. Histopathology was the reference standard. The primary analysis defined cancer as International Society of Urological Pathology (ISUP) grade group ≥ 1 (Gleason score ≥ 6); sensitivity analysis defined clinically significant cancer as ISUP grade group ≥ 2. STATISTICAL TESTS: Generalized Estimating Equations were used for clustered data. Receiver operating characteristic (ROC) analysis with the Obuchowski-Rockette model was used to compare the area under the ROC curve (AUC). Cohen's κ assessed inter-reader agreement; two-sided p < 0.05 indicated significance. RESULTS: For ISUP ≥ 1, uAI increased novice-level/experienced-reader AUCs (0.684-0.744; 0.757-0.794). At PI-RADS ≥ 3, novice-level sensitivity/specificity significantly improved (0.71-0.79; 0.46-0.58). Experienced-reader sensitivity gains were nonsignificant (p = 0.344/0.291). For ISUP ≥ 2 at ≥ 3, all-reader sensitivity/specificity increased (0.76-0.82; 0.47-0.57). Novice-level κ increased at ≥ 3/≥ 4 (0.582-0.700; 0.654-0.741). DATA CONCLUSION: uAI assistance improved diagnostic performance, with multi-metric improvements in novice-level readers. TECHNICAL EFFICACY: Stage 3.
BACKGROUND: Deep learning (DL) methods have shown potential for predicting clinically significant prostate cancer (csPCa), but radiologists often face challenges in effectively leveraging these techniques for csPCa predi...BACKGROUND: Deep learning (DL) methods have shown potential for predicting clinically significant prostate cancer (csPCa), but radiologists often face challenges in effectively leveraging these techniques for csPCa prediction. PURPOSE: To develop an automated DL model based on biparametric-MRI (bpMRI) and propose a human-machine collaborative strategy for predicting csPCa. STUDY TYPE: Retrospective. POPULATION: A total of 4305 patients were enrolled. Centers 1-2 and 4-7 comprised the training (2437 patients, mean age 68 ± 8) and the internal validation (581 patients, mean age 67 ± 8) cohorts; Centers 8-10 comprised the external validation cohort 1 (622 patients, mean age 71 ± 8), and Center 3 comprised the external validation cohort 2 (665 patients, age not available). FIELD STRENGTH/SEQUENCE: T2-weighted imaging (T2WI) using fast or turbo spin echo and diffusion-weighted imaging (DWI) using single-shot echo planar imaging were acquired at 1.5 and 3 T. ASSESSMENT: A DL model (UFormer) including prostate segmentation and csPCa prediction was constructed using bpMRI. Its performance was evaluated in two external validation cohorts (EVCs) and compared with that of radiologists. Further, a UFormer-radiologist collaborative predictive strategy was proposed. STATISTICAL TESTS: Area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, DeLong test, and McNemar test. p < 0.05 was considered significant. RESULTS: Compared with radiologists' Prostate Imaging Reporting and Data System (PI-RADS) assessment, UFormer-combined radiologists showed significantly higher AUC and accuracy of 0.918 ± 0.012 and 0.857 ± 0.014 for the less-experienced radiologists, 0.931 ± 0.010 and 0.870 ± 0.014 for the more-experienced radiologists, respectively, due to greatly increasing specificity by 121.7% for the less-experienced radiologists and 60.2% for the more-experienced radiologists in EVC1. Additionally, UFormer identified 86.5% and 93.9% of non-csPCa patients, who had been interpreted originally as PI-RADS 3 by more- and less-experienced radiologists, respectively. DATA CONCLUSIONS: UFormer enhanced the predictive performance of radiologists and narrowed performance gaps between experience levels. The UFormer-radiologist collaborative paradigm combined model advantages with PI-RADS assessment, providing a strategy for clinical application. EVIDENCE LEVEL: 4. TECHNICAL EFFICACY: Stage 2.
BACKGROUND: Timely identification of early recurrence (≤ 6 months) may improve prognosis of glioblastoma (GBM), but conventional MRI has shown limited accuracy in this setting. PURPOSE: Risk-assessment models and a nomog...BACKGROUND: Timely identification of early recurrence (≤ 6 months) may improve prognosis of glioblastoma (GBM), but conventional MRI has shown limited accuracy in this setting. PURPOSE: Risk-assessment models and a nomogram were constructed by integrating SyMRI metrics, clinical-pathological variables, and cMRI features, including contrast-enhanced T1-weighted imaging and fluid-attenuated inversion recovery (FLAIR) findings. STUDY TYPE: Retrospective observational study. POPULATION: Seventy-eight patients with GBM (median age, 59 years; 44 [56.4%] males). FIELD STRENGTH/SEQUENCE: 3 T; pre- and post-contrast three-dimensional T1-weighted imaging, FLAIR, and SyMRI. ASSESSMENT: Histogram-based quantitative metrics were extracted from SyMRI maps using subregions defined on fused FLAIR and contrast-enhanced T1-weighted images. All candidate clinical-pathological variables, cMRI features, and SyMRI-derived metrics were entered directly into least absolute shrinkage and selection operator (LASSO) regression for variable selection. Risk-assessment models and a nomogram were constructed. STATISTICAL TESTS: Multivariable logistic regression, multicollinearity assessment using variance inflation factors, receiver operating characteristic curve analysis with DeLong test, stratified 10-fold cross-validation, leave-one-out cross-validation, nested 5-fold cross-validation, calibration curves, and decision curve analysis. A two-sided p < 0.05 was considered statistically significant. RESULTS: Multivariate analysis identified reduced T2 entropy (< 2.113) in enhancement-corresponding regions (odds ratio [OR] = 0.08), thick linear or nodular residual cavity wall enhancement (OR = 5.28), corpus callosum involvement (OR = 5.08), and O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation (OR = 0.19) as variables associated with early recurrence. The integrated model achieved the highest performance (AUC = 0.864). Nested cross-validation showed moderate internal validation performance, with an AUC of 0.724 (0.722-0.726). DATA CONCLUSION: Histogram-based pre-radiotherapy SyMRI metrics, particularly T2 entropy, were associated with early GBM recurrence. The integrated model achieved the highest apparent performance; nested internal validation showed moderate performance. External validation in larger multicenter cohorts is required before clinical implementation. EVIDENCE LEVEL: 3. TECHNICAL EFFICACY: Stage 2.
BACKGROUND: Existing strategies for free-breathing cine cardiac MR, including real-time imaging, struggle with complex arrhythmias such as atrial fibrillation (AF). Dynamic regularized adaptive clustering optimization (D...BACKGROUND: Existing strategies for free-breathing cine cardiac MR, including real-time imaging, struggle with complex arrhythmias such as atrial fibrillation (AF). Dynamic regularized adaptive clustering optimization (DRACO) has been proposed to generate high-quality and quantifiable breath-held cine images for patients in AF. PURPOSE: To tailor DRACO for free-breathing and evaluate its potential to simultaneously handle cardiac and respiratory motion. STUDY TYPE: Prospective. SUBJECTS: 10 sinus rhythm (10 males), 20 AF (18 males). FIELD STRENGTH/SEQUENCE: 3.0 T, balanced steady-state free precession (bSSFP) with sorted golden-step, ECG-gated segmented cine, and real-time. ASSESSMENT: Patients were imaged using the golden-step sequence under breath-held and free-breathing conditions, and the reference bSSFP cine sequence. Signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), edge sharpness, and image quality (4-point Likert scale) were assessed. STATISTICAL TESTS: Paired t-test, Friedman test, regression analysis, Fleiss' Kappa. Significance level p < 0.05. RESULTS: The SNR and CNR values of DRACO images showed no significant differences between breath-held and free-breathing conditions in both sinus rhythm (SNR p = 0.53; CNR p = 0.87) and in AF (SNR p = 0.56; CNR p = 0.55). Image sharpness was not significantly different between breath-held and free-breathing in both sinus rhythm (p = 0.24) and AF (p = 0.79). In AF patients, breath-held DRACO and free-breathing DRACO had significantly higher image quality scores than real-time cine. In sinus rhythm patients, there was no significant difference (p = 0.13) between breath-held or free-breathing DRACO and segmented cine. Despite significant within-patient ejection fraction (EF) differences between breath-held versus free-breathing in 80% of sinus rhythm and 70% of AF patients, the average EF between breath-held versus free-breathing DRACO showed high correlation (sinus rhythm R = 0.71; AF R = 0.71), demonstrating DRACO's robustness to motion variation. DATA CONCLUSION: DRACO is robust to respiratory motion and AF-related irregular cardiac motion. It enables high-quality cine image generation in both regular and irregular cardiac motion, with or without breath holding. TECHNICAL EFFICACY: Stage 2.