Radiology. Imaging cancer最新文献

Purpose To quantify functional liver volume (FLV) loss relative to minimal ablative margins (MAMs) achieved during percutaneous thermal ablation (TA) of liver tumors in COVER-ALL randomized controlled trial (RCT) participants. Materials and Methods In the COVER-ALL single-center RCT (June 2020 through October 2023), software-based ablation confirmation (AC) (experimental) and visual margin (control) assessment were compared among participants with histology-agnostic liver tumors. In this post hoc analysis, the authors compared CT-derived volumes, MAMs, and changes in laboratory-based hepatic function using Wilcoxon rank sum and Spearman correlation tests. Results Among 100 participants (mean age ± SD, 57.8 years ± 13.2; 61 male; experimental group, n = 74, 98 tumors ablated, median diameter = 1.5 cm; control group, n = 26, 41 tumors ablated, median diameter = 1.3 cm), the experimental and control group baseline median FLVs (1707 cm³ [IQR, 1467-1964 cm³] vs 1722 cm³ [IQR, 1338-1916 cm³]; P = .84) were comparable. The median MAM was larger in the experimental versus the control group (6 mm [IQR, 4.5-7.9 mm] vs 1 mm [IQR, 0-4 mm]; P < .001). The median percentage FLV loss was larger in the experimental versus the control group (2.9% [IQR, 1.8%-4.4%] vs 1.7% [IQR, 1.2%-3.4%]; P = .03). The median postablation FLV was comparable between the experimental and control groups (1643 cm³ [IQR, 1403-1886 cm³] and 1696 cm³ [IQR, 1315-1843 cm³]; P = .87). We observed no association between percentage FLV loss and changes in serum albumin (ρ = -0.153, P = .16) or total bilirubin concentrations (ρ = -0.128, P = .25). Conclusion Liver tumor TA resulted in minimal percentage FLV loss. Software-based AC use increased the MAM at the expense of a negligible increase in percentage FLV loss. Keywords: Ablation Techniques, Segmentation, Liver, Volume Analysis ClinicalTrials.gov: NCT04083378 Supplemental material is available online. © RSNA, 2026.

Purpose To characterize lipidomic profiles in the VX2 rabbit tumor model and assess potential lipid plasticity across tumor microenvironments in the liver and flank. Materials and Methods VX2 tumors were implanted into the liver in four rabbits and both flanks in two rabbits. After 10-14 days, tumor-containing tissues were harvested, snap frozen, sectioned, and analyzed using matrix-assisted laser desorption ionization mass spectrometry imaging (MALDI-MSI) in both positive and negative ion modes. Tandem mass spectrometry was used for further characterization. Lipid identities were determined based on accurate mass, fragmentation patterns, and database matching. Molecular ion distributions were spatially correlated with hematoxylin-eosin-stained sections. Results MALDI-MSI revealed distinct lipidomic profiles between liver and flank tumors. Oleate (mass-to-charge ratio [m/z] 281.24) was consistently localized to viable tumor regions in both sites. Phosphatidic acid 18:0/18:1(m/z 701.51) was detected in only liver tumors, whereas phosphatidylinositol 18:2/18:0 (m/z 861.55) was prominent in flank tumors but absent from liver tumors, despite being present in surrounding healthy parenchyma. Phosphatidylcholine (PC) species, notably PC 34:1 (m/z 798.52) and PC 36:2 (m/z 808.56), and PC (P-36:2) (m/z 824.56), showed heterogeneous spatial distributions across both anatomic sites. Conclusion MALDI-MSI enabled spatial mapping of lipid distributions within VX2 tumors, revealing both shared and site-specific alterations. These findings reveal heterogeneous spatial lipid distributions in the VX2 model and may reflect microenvironment-driven lipid plasticity. Keywords: Mass Spectrometry Imaging, VX2 Rabbit Tumor Model, Lipidomic Profiles, Ablation Techniques © RSNA 2026.

Purpose To assess whether k-means clustering (KMC) analysis of ultrafast dynamic contrast-enhanced (DCE) MRI kinetics can help differentiate ductal carcinoma in situ (DCIS) grades and identify aggressive lesions. Materials and Methods In this retrospective study of patients with DCIS who underwent ultrafast DCE MRI (3.0-9.0 seconds per image) between May 2015 and September 2024, KMC was performed to classify the breast parenchyma, including lesion voxels, into five clusters regarding maximum slope (A × α, where A is initial contrast agent uptake upper limit and α the contrast agent uptake rate). The model-derived and physiologic parameters were computed for each cluster's mean curve and compared with weighted averages from clusters 3-5 across DCIS grades. The Fisher exact, Kruskal-Wallis, Cuzick, and Wilcoxon rank sum tests were performed to identify optimal classifiers for identifying low-grade DCIS and receiver operating characteristic curve analysis to evaluate parameter performance in differentiating between low-grade and intermediate- to high-grade DCIS. Results Among 57 female patients (median age, 52 years; IQR, 20 years; 59 affected breasts), α was associated with different DCIS grades (median α: 2.95 [IQR, 3.01] for low; 5.07 [IQR, 3.63] for intermediate; 6.37 [IQR, 6.33] for high; P = .04, Kruskal-Wallis). The MRI parameters α, A × α, area under the enhancement curve for the first 30 seconds, influx transcapillary transfer constant (K^trans), and efflux transcapillary rate constant (K_ep) had areas under the receiver operating characteristic curve greater than 0.70, with α achieving the highest area under the receiver operating characteristic curve (0.77; 95% CI: 0.56, 0.92; P = .02) for identifying low-grade DCIS. Conclusion KMC-derived kinetic parameters from ultrafast DCE MRI could differentiate low-grade from intermediate- to high-grade DCIS, with α showing the best performance. Keywords: k-Means Clustering, Ultrafast Dynamic Contrast-enhanced MRI, Ductal Carcinoma in situ Grading © RSNA, 2026.

Purpose To develop a fully automated hybrid approach to predict sacral tumor types from preoperative noncontrast CT (NCCT ) images. Materials and Methods In this retrospective, multicenter study, scans were available in 690 patients who had histopathologically confirmed preoperative sacral NCCT performed between January 2011 and May 2024. A fully automated hybrid model integrated two deep convolutional neural network models (model 1 and model 2) through a fully automated pipeline. Model 1 segments tumors and hip bones automatically from NCCT images, producing masks that are used by model 2. For the first time, the hip bone was used as a reference frame for tumor localization. The second model, CL-MedImageNet, is an innovative six-classification model that allows the simultaneous input of tumor images, clinical data, and location information. This streamlined, automated system ensures efficient data integration and processing between the two models. The efficacy of the model was assessed in comparison to that of radiologists, using metrics including the area under the curve (AUC), F1 score, and confusion matrix. Results In all, 690 patients (mean age, 46 years ± 17 [SD]; 377 male patients) were included. Segmentation achieved mean Dice coefficients of 0.82 ± 0.11 (validation), 0.81 ± 0.12 (internal test), and 0.81 ± 0.12 (external test) after postprocessing; interobserver Dice coefficient was 0.96. The CL-MedImageNet classifier attained macro average AUCs of 0.89 (95% CI: 0.83, 0.93), 0.88 (95% CI: 0.84, 0.92), and 0.87 (95% CI: 0.79, 0.92) in validation, internal, and external test sets, respectively, with macro average F1 scores of 0.63, 0.63, and 0.56. The highest achieved precision and sensitivity were both 0.66 across all sets. CL-MedImageNet outperformed radiologists (macro average AUCs, 0.87 vs 0.80, P = .002; 0.87 vs 0.83, P = .45). Conclusion The fully automated NCCT-based CL-MedImageNet pipeline demonstrated high segmentation accuracy and robust six-class classification, outperforming expert radiologists. Keywords: Applications - CT, Deep Learning, Radiomics, Segmentation, Skeletal-Axial, Pelvis, Sacral Tumors Supplemental material is available for this article. © The Author(s) 2026. Published by the Radiological Society of North America under a CC BY 4.0 license.

Purpose To develop a radiomics model based on hepatobiliary phase gadolinium ethoxybenzyl-diethylenetriaminepentaacetic acid (EOB)-enhanced MRI features at the tumor margin to predict microvascular invasion in high-risk solitary hepatocellular carcinoma (HR-sHCC), determine the optimal margin region, and explore the underlying biologic mechanisms. Materials and Methods This retrospective study included patients with HR-sHCC from three medical centers between April 2015 and December 2022. Radiomics features were extracted from 121 volumes of interest (VOIs) at the tumor margin at EOB MRI. Nine combinations of statistical and machine learning methods were used to construct and validate the optimal margin region-based radiomics model. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), and patient stratification was evaluated with Kaplan-Meier and log-rank analyses. RNA sequencing data underwent differential expression analysis with DESeq2, followed by Kyoto Encyclopedia of Genes and Genomes (ie, KEGG) and Gene Ontology (ie, GO) enrichment, and immune cell infiltration was assessed using xCell and EPIC. Results A total of 436 patients (mean age, 57.7 years ± 8.8 [SD]; 352 male) were included: 254 in the training, 108 in the internal test, and 74 in the external test cohorts. Receiver operating characteristic analysis showed AUCs of 0.80 (95% CI: 0.74, 0.86), 0.76 (95% CI: 0.66, 0.85), and 0.72 (95% CI: 0.58, 0.86), respectively. The model effectively stratified patients by overall and disease-free survival (all P < .05). RNA sequencing revealed extracellular matrix remodeling, transforming growth factor-β signaling, and M2 macrophage infiltration in high optimal margin region-score tumors. Conclusion The optimal margin region-based radiomics model, derived from EOB MRI, effectively captured tumor margin heterogeneity. Keywords: MRI, Machine Learning, Radiomics, Radiogenomics, Abdomen/GI, Liver, Surgery, High-Risk Solitary Hepatocellular Carcinoma, Tumor Margin, Microvascular Invasion, Gd-EOB-DTPA-enhanced MRI, OATP1B3 © The Author(s) 2026. Published by the Radiological Society of North America under a CC BY 4.0 license. Supplemental material is available for this article.