Radiology最新文献

Background Volumetric evaluation of liver and spleen size is useful, yet pediatric reference values remain limited and lack external validation. Purpose To establish normative liver and spleen volumes from contrast-enhanced CT in children and validate them in external regional and international datasets. Materials and Methods In this retrospective study, contrast-enhanced abdominal CT scans (patient age, 2-18 years) from Asan Medical Center and Pusan National University Yangsan Hospital (January 2005 to June 2021) were collected for the derivation dataset. Patients with chronic diseases, liver and/or spleen imaging abnormalities, or missing height and/or weight measurement within 3 months were excluded. Portal-phase CT images were segmented automatically with manual correction. Data from Seoul National University Children's Hospital (January 2018 to December 2021) and Cincinnati Children's Hospital Medical Center (January 2018 to July 2021) were used for regional and international validation, respectively. Model performance was assessed with the average pinball loss and absolute errors for predicting 50th percentiles. Results A total of 1298 children (median age, 11.3 years [IQR, 8.0-15.0 years]; 660 female) were included (derivation set, n = 1030 [all Asian]; regional validation set, n = 114 [all Asian]; international validation set, n = 154 [racially diverse]). The model including sex, height, and weight had the lowest average pinball loss for predicting 50th percentile liver (54.50 mL) and spleen (18.64 mL) volume and was therefore selected. Compared with previously published formulas, the model showed lower average pinball loss in the regional validation dataset for both liver (5th, 11.55 mL vs 11.93 mL; 50th, 50.88 mL vs 55.47-116.02 mL; 95th, 15.85 mL vs 19.49 mL) and spleen volumes (5th, 3.69 mL vs 3.74 mL; 50th, 15.19 mL vs 16.03 mL; 95th, 5.01 mL vs 5.59 mL). In international external validation, the average pinball loss was lower for the 50th percentiles (liver, 50.28 mL vs 50.30 mL; spleen, 20.97 mL vs 21.42 mL), and absolute errors were comparable (liver, 100.56 mL vs 100.60 mL [P = .53]; spleen, 41.93 mL vs 42.83 mL [P = .69]). Conclusion Normative liver and spleen volumes in children derived using quantile regression models and the respective predictive performance were presented. © RSNA, 2026 Supplemental material is available for this article.

Background Accurate preoperative identification of pathologic extranodal extension (ENE) at CT is essential for precise treatment decisions in laryngeal and hypopharyngeal squamous cell cancer (LHSCC). However, human interpretation of ENE is neither reliable nor reproducible. Purpose To develop and evaluate the diagnostic performance of a new deep learning tool, DeepENE, in detecting metastatic and ENE lymph nodes on preoperative CT scans in patients with LHSCC in a multicenter cohort. Materials and Methods In this retrospective study, patients with LHSCC from Zhongshan Hospital, Fudan University (April 2011-August 2022), were included in training, validation, and internal test sets to develop DeepENE. For the reference standard, lymph nodes were segmented on CT scans and labeled for metastasis and ENE status based on pathologic findings. DeepENE was tested using three external cohorts of patients with LHSCC (external test sets 1-3) and one external cohort of patients with oral squamous cell carcinoma. The primary diagnostic metric was the area under the receiver operating characteristic curve (AUC). The performance of DeepENE was compared with that of five board-certified head and neck cancer specialists using the DeLong method. Results Overall, 289 patients with LHSCC with 1954 pathologically confirmed lymph nodes were evaluated. DeepENE achieved an AUC of 0.93 for ENE diagnosis in the internal test set under fivefold cross-validation, and AUCs of 0.96, 0.87, and 0.90 in external test sets 1, 2, and 3, respectively. DeepENE outperformed the five experts, especially in early-stage ENE detection in external test set 2 (AUC of 0.87 for DeepENE vs mean AUC of 0.66 for readers; P < .001). In external test set 1, DeepENE maintained a high sensitivity of 97% at specificity of 90%, compared with experts' mean sensitivity of 77% (P = .003). In external test sets 2 and 3, DeepENE had sensitivity of 78% and 80%, compared with experts' mean sensitivity of 36% (P < .001) and 46% (P < .001), respectively. Conclusion DeepENE accurately detected ENE on preoperative CT scans in patients with LHSCC and outperformed head and neck cancer specialists. © RSNA, 2026 Supplemental material is available for this article.

Background Dysfunction of the peripheral nerve microcirculation is crucial in the onset and progression of diabetic neuropathy. Purpose To evaluate the feasibility of superresolution US for quantifying tibial nerve microvasculature in diabetic adults with and without peripheral neuropathy, and in control participants. Materials and Methods In this prospective single-center study, consecutive participants were enrolled between June 2024 and October 2024 and divided into three groups: group I, diabetic peripheral neuropathy with ulcers; group II, type 2 diabetes without peripheral neuropathy; and group III, the control group. All participants underwent conventional US in the tibial nerves followed by superresolution US with intravenous microbubble injection. Regions of interest were delineated manually along the tibial nerve. Quantitative parameters, including interfascicular and intrafascicular vessel ratio, complexity, density, velocity, perfusion index, and single-vessel curvature were compared using one-way analysis of variance (parametric) and Dunn-corrected Kruskal-Wallis (nonparametric) tests. Results A total of 100 participants were enrolled, including 30 participants in group I (mean age, 67 years ± 10 [SD]; 23 men), 35 in group II (mean age, 61 years ± 11; 20 men), and 35 in group III (mean age, 54 years ± 14; 15 men). Superresolution US showed group I had higher vessel ratio (23% vs 10% vs 6%), complexity level (1.3 vs 1.2 vs 1.1), maximum density (25 vs 20 vs 12), mean velocity (12.6 vs 8.7 vs 8.4 mm/sec), perfusion index (4.5 vs 1.4 vs 0.9), and curvature (1.19 vs 1.09 vs 1.05) compared with groups II and III, respectively (all P < .05). Microvascular tortuosity was observed in 83% (25 of 30), 43% (15 of 35), and 14% (five of 35) of groups I, II, and III, respectively (P < .001). Conclusion Superresolution US provided a 10-μm scale resolution view of tibial nerve microvascular structure, helping to identify microvascular alterations in participants with diabetic peripheral neuropathy with ulcers. © RSNA, 2026 Supplemental material is available for this article.

The Dixon method is a fat suppression method based on the separation of fat and water signals. Originally limited to gradient-echo sequences, it can now be integrated with fast/turbo spin-echo MRI sequences, unlocking its potential for musculoskeletal imaging. Key advantages include robust fat suppression and the ability to generate four image contrasts from a single acquisition-including fat-only images, which are specific to the signal of fat-thereby providing opportunities for protocol optimization. Additionally, quantitative data readily available from Dixon sequences provide information on intralesional fat, which may serve as a diagnostic aid complementing the morphologic assessment. The Dixon method has been increasingly used in musculoskeletal MRI, primarily in bone marrow imaging to help detect and characterize focal marrow lesions, vertebral compression fractures, and degenerative spine conditions. Other applications include whole-body imaging, imaging of rheumatologic and neuromuscular disorders, tumor characterization, and imaging near metallic implants. Limitations include fat-water swapping artifacts and unreliable fat quantification (eg, when mineralization is increased). This article covers the strengths, applications, and limitations of the Dixon method in musculoskeletal MRI, with practical tips for incorporating this method into routine protocols, highlighting its most effective applications and noting areas where it is less reliable.

Background Chronic subdural hematomas (cSDHs) are associated with high recurrence risks following surgical evacuation. The EMBOLISE trial demonstrated that, compared with surgery alone, adjunctive middle meningeal artery embolization (MMAE) significantly reduced reoperation rates. However, given the limitations of the clinical end points of the trial, which may be subject to interrater variability and certain biases, the quantitative imaging metrics need to be evaluated. Purpose To evaluate the prespecified imaging end points of the EMBOLISE trial and assess the long-term resolution of cSDH through quantitative imaging analyses. Materials and Methods EMBOLISE was a multicenter, randomized, interventional trial conducted across 39 U.S. sites between December 2020 and August 2023. Prespecified secondary imaging end points included changes in hematoma volume and thickness and midline shifts from 24 hours to 90 days after the procedure at CT and MRI. The post hoc analyses performed herein extended the assessment to 180 days and included absolute hematoma metrics. Mixed-effects modeling was employed to adjust for confounders. Results Four hundred patients were enrolled in the EMBOLISE study, among whom 352 were included (mean age, 72 years ± 10.4 [SD]; 256 men). The mean cSDH volume was 126 mL at screening, with no intergroup differences. At 90 and 180 days, the MMAE plus surgery group had lower cSDH volumes (20.6 mL vs 28.9 mL [P = .03] and 19.4 mL vs 31.5 mL [P = .04], respectively). Mixed-effects models revealed a 6.9 mL (95% CI: -13.5, -0.40; approximately 25%) greater volume reduction and an 8.4 mL (95% CI: -15.2, -1.6; approximately 30%) lower absolute volume at 90 days in the MMAE group There was no evidence of a difference in the prespecified secondary imaging end points between the groups. Conclusion While the prespecified secondary imaging end points did not significantly differ, the absolute 90- and 180-day hematoma volumes were significantly lower in patients who received MMAE and surgery. Confounder-adjusted mixed-effects analysis indicated a greater reduction in hematoma volume with adjunctive MMAE. ClinicalTrials.gov identifier NCT04402632 © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Ramasamy and Baker in this issue.

Background The 2023 update of the American Association for the Study of Liver Diseases (AASLD version 2023 [hereafter, v2023]) guidance introduced new triggers for hepatocellular carcinoma (HCC) surveillance, incorporating US visualization score, increasing α-fetoprotein (AFP) level, and lesion growth, but has not yet been validated. Purpose To assess the surveillance performance of AASLD v2023 for HCC detection in comparison to Liver Imaging Reporting and Data System (LI-RADS) version 2017 (hereafter, v2017) and AASLD version 2018 (hereafter, v2018) in participants at high risk of HCC. Materials and Methods This prospective study consecutively enrolled high-risk participants undergoing US and AFP surveillance across three institutions between July 2023 and October 2024. Surveillance performance of US LI-RADS v2017, AASLD v2018, and AASLD v2023 for HCC detection was compared through sensitivity, specificity, positive predictive value, and negative predictive value (NPV). Multivariable logistic regression helped identify predictors of false-negative and false-positive classifications. Results Among 953 participants (median age, 51 years; IQR, 45-58 years; 658 men), 5% (50 of 953) had HCC. The new surveillance triggers (visualization score VIS-C, increasing AFP level, and lesion growth) demonstrated sensitivities ranging from 8% (four of 50 participants) to 48% (24 of 50 participants), with high specificities of 94% (849 of 903 participants) to 99.4% (898 of 903 participants). As an integrated algorithm, AASLD v2023 achieved a sensitivity of 94% (47 of 50 participants) and NPV of 99.6% (758 of 761 participants) for HCC detection, surpassing US LI-RADS v2017 (sensitivity, 60% [30 of 50 participants], P < .001; NPV, 98% [814 of 834 participants], P = .001) and AASLD v2018 (sensitivity, 76% [38 of 50 participants], P = .02; NPV, 98.5% [805 of 817 participants], P = .04). Specificity was lower than that with US LI-RADS v2017 (84% [758 of 903 participants] vs 90% [814 of 903 participants], P < .001) and AASLD v2018 (84% [758 of 903 participants] vs 89% [805 of 903 participants], P = .001). For early HCC, sensitivity remained superior to that of other algorithms (P < .001, P = .04). At multivariable analysis, AFP level of less than 20 ng/mL (odds ratio, 11.76; P < .001) and absence of cirrhosis (odds ratio, 2.45; P = .03) were independently associated with false-positive findings for AASLD v2023. Conclusion AASLD v2023 improved the sensitivity and NPV for HCC surveillance, outperforming US LI-RADS v2017 and AASLD v2018. Chinese Clinical Trial Registry no. ChiCTR2100054330 © RSNA, 2026 Supplemental material is available for this article.