Background Multiple myeloma (MM) imaging is currently suboptimal, which may limit patient care. Zirconium 89 ([89]Zr) deferoxamine (DFO)-daratumumab is a novel CD38-targeted immuno-PET tracer. Purpose To compare [89]Zr-DFO-daratumumab PET/CT against standard-of-care fluorine 18 ([18]F) fluorodeoxyglucose (FDG) PET/CT for detection of MM. Materials and Methods Participants with MM were accrued to this prospective phase 2 trial between March 2022 and July 2025. Standard-of-care [18]F-FDG PET/CT and bone marrow biopsy were performed, along with research [89]Zr-DFO-daratumumab PET/CT. If there were discrepant lesions between the two scans, image-guided biopsy was considered as the reference standard. The McNemar test was used to compare disease detection between the two imaging modalities. Results This study included 60 participants (median age, 63 years [IQR, 57-71 years]; 32 women). [89]Zr-DFO-daratumumab PET/CT demonstrated at least one positive lesion in 43% (26 of 60; 95% CI: 31, 57) of participants, and [18]F-FDG PET/CT demonstrated at least one positive lesion in 47% (28 of 60; 95% CI: 33, 60) of participants. While there was no overall difference in disease detection between the two imaging modalities (P = .81), MM was demonstrated only with [89]Zr-DFO-daratumumab PET/CT in 13% (eight of 60) of participants, with subsequent biopsy confirmation of active myeloma. Conclusion In participants with MM, CD38-targeted [89]Zr-DFO-daratumumab immuno-PET helped detect active myeloma in some participants with no signs of disease at [18]F-FDG PET/CT. Clinical trial registration no. NCT04814615 © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Choyke in this issue.

Background The incidence of lymph node (LN) metastasis in patients with part-solid nodules (PSNs) is known to be low, which makes the intraoperative evaluation of LNs a controversial issue. Purpose To investigate the incidence of LN metastasis and explore clinical-radiologic predicting factors for LN metastasis in PSN lung adenocarcinomas with a solid portion diameter of 30.0 mm or less in a large surgical study sample. Materials and Methods This retrospective study included patients with part-solid lung adenocarcinomas with solid portion diameters of 30.0 mm or less on preoperative CT images from July 2023 to December 2024. For each PSN, overall tumor size, solid portion diameter, consolidation-to-tumor ratio (CTR), location, margin, shape, spiculation, lobulation, pleural retraction, bubble-like lucency, and air bronchogram were independently evaluated by two chest radiologists blinded to clinical information and pathology results. Univariable and multivariable logistic regression analyses were performed. Diagnostic performance was evaluated using receiver operating characteristic curve, calibration, and decision curve analysis. Results This study included 1444 patients (median age, 59 years; IQR, 51-67 years; 931 women). The frequency of LN metastasis was 4.7% (68 of 1444 patients). None of the 115 patients with a CTR of 0.25 or less had LN metastasis (N1: 0 of 115; N2: 0 of 115). Rarely (0.2%, one of 552), LN metastasis was found in patients with a solid portion diameter of 6.0 mm or less. Solid portion diameter was the only independent predictor of LN metastasis (odds ratio, 1.25; 95% CI: 1.20, 1.31) in multivariable analysis and showed a superior diagnostic performance compared with CTR (area under the receiver operating characteristic curve, 0.86 [95% CI: 0.82, 0.90] vs 0.75 [95% CI: 0.68, 0.80]; P < .001), along with better calibration (Brier score, 0.039 vs 0.044) and greater net benefit. The clinical actionable cutoff for solid portion diameter was 7.3 mm (sensitivity, 95.6%; specificity, 49.6%). Conclusion Solid portion diameter had a superior diagnostic performance compared with CTR in predicting LN metastasis in patients with part-solid lung adenocarcinomas of solid portion diameter of 30.0 mm or less at CT. © RSNA, 2026 Supplemental material is available for this article.

Background Microwave ablation (MWA) combined with ethanol injection (EI) is a minimally invasive technique under preliminary investigation for the treatment of nodular retrosternal goiter (RSG) and has shown promising treatment efficacy. However, a comparative analysis between ablation and surgery is warranted to draw definitive conclusions. Purpose To compare the long-term efficacy and safety of MWA combined with EI versus surgery for patients with nodular RSG. Materials and Methods This retrospective study included patients with nodular RSG treated with MWA combined with EI or surgery between October 2016 and June 2024. Propensity score matching was performed for patients in the ablation and surgery groups, followed by comparisons between the two groups with independent-sample t test, Mann-Whitney U test, χ[2], or Fisher exact test. Primary outcomes were treatment success, symptom score, and complication rate. Secondary outcomes were cosmetic score, hospitalization, procedure time, estimated blood loss, incision length, and cost. Results This study included 555 patients (mean age ± SD, 51 years ± 14.4; 414 women). After propensity score matching, 224 patients in the ablation group and 224 patients in the surgery group were observed for a median of 48 months (IQR, 24-60 months). No evidence of a difference was observed in the treatment success rates between the ablation and surgery group (98.7% [221 of 224] vs 100% [224 of 224], respectively; P = .08), while both modalities effectively lowered symptom scores from baseline (both P < .001). Compared with surgery, MWA resulted in a lower complication rate (2.2% [five of 224] vs 34.8% [78 of 224], P < .001), shorter posttreatment hospitalization (P < .001), shorter procedure time (P < .001), reduced estimated blood loss (P < .001), shorter incision length (P < .001), lower cost (P < .001), and lower cosmetic score (mean, 1.1 ± 0.3 vs 1.2 ± 0.4; P = .03) at the 12-month follow-up. Conclusion MWA combined with EI was as effective as surgery for nodular RSGs, providing comparable symptom relief while demonstrating fewer complications, offering a safer alternative for eligible patients. © RSNA, 2026 Supplemental material is available for this article.

Background Contrast-enhanced mammography (CEM) is increasingly used in breast cancer diagnostics; however, randomized trials assessing the added value of CEM in preoperative staging of breast cancer beyond digital mammography, digital breast tomosynthesis, and US are lacking. Purpose To assess the added value of CEM for preoperative breast cancer staging in a randomized controlled trial. Materials and Methods This prospective randomized controlled trial (Added Value of Preoperative Contrast Enhanced Mammography in Staging of Malignant Breast Lesions) enrolled women diagnosed with early-stage breast cancer eligible for primary surgery at two breast centers in Sweden. Participants were randomized 1:1 to preoperative CEM or control from November 2020 through April 2024. The primary outcome was a 10% change in treatment due to preoperative CEM findings. Secondary outcomes included differences in reoperation and mastectomy rates between groups and accuracy of malignant extent estimation with CEM, mammography/tomosynthesis, and US using histopathologic extent as the reference, assessed with Bland-Altman statistics and concordance analysis. Subgroup analyses were performed according to menopausal status, breast density, histopathologic type of cancer, lesion size, and presence of microcalcifications at baseline imaging. Results In total, 415 participants (median age, 66 years; IQR, 56-73 years) completed the trial: 202 in the CEM group and 213 in the control group. Treatment was changed in 17 of 202 participants (8.4%; 95% CI: 5.3, 13.1) based on CEM findings. No evidence of differences was observed between the CEM and control groups with regard to mastectomy (17% [35 of 202] vs 18% [39 of 213]; P = .78) or reoperation (7.9% [16 of 202] vs 10% [22 of 213]; P = .39) rates. CEM showed the closest agreement with histopathologic malignant extent (bias, 3.5 mm), outperforming mammography and/or tomosynthesis (4.8 mm) and US (10.7 mm) overall and in nine of 10 predefined subgroups, except for ductal carcinoma in situ. Conclusion This trial did not meet its primary end point of a 10% treatment change; however, preoperative CEM improved malignant extent estimation. ClinicalTrials.gov: NCT04437602 © 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. See also the editorial by Mesurolle and El Khoury in this issue.

Pulmonary hypertension (PH) is a progressive disease associated with high morbidity and mortality. PH is classified into five groups based on underlying pathophysiology and etiology: pulmonary arterial hypertension (group 1); PH due to left heart disease (group 2); PH secondary to lung disease (group 3); pulmonary artery obstruction, including chronic thromboembolic pulmonary hypertension (CTEPH) (group 4); and PH due to unclear or multifactorial mechanisms (group 5). Radiologists play a major role in the multidisciplinary evaluation of patients with PH alongside specialists in pathology, pulmonary medicine, and cardiology. Given the widespread use of CT, imaging may be the first modality to suggest PH and plays a central role in the diagnosis and management of CTEPH. This review describes the pathophysiology of the five PH groups and emphasizes imaging findings that raise suspicion for PH. Discussion includes the roles of CT and MRI in PH evaluation, risk stratification, and assessment of treatment response, as well as the current applications of multienergy CT, echocardiography, and cardiac MRI in PH management. Keywords: CT Angiography, CT-Dual Energy, CT, MR Imaging, Pulmonary Arteries, Pulmonary, Echocardiography, Thorax, Vascular, Pathology, Pulmonary Hypertension, Chronic Thromboembolic Disease, Chronic Thromboembolic Pulmonary Hypertension, Pulmonary Embolism, Pulmonary Arterial Hypertension, Pulmonary Venoocclusive Disease, Pulmonary Capillary Hemangiomatosis, Pulmonary Hypertension Groups, Dual-Energy CT, Photon-counting CT, Spectral CT © RSNA, 2026.

Cardiac myxomas are the second most common benign cardiac tumor in adults. Most arise from the interatrial septum in the left atrium. Although most patients present with cardiac or embolic symptoms, some lesions are detected incidentally at imaging performed for unrelated indications. This case report describes the incidental discovery of a right atrial mass partially visualized at CT of the abdomen and pelvis during workup for cholecystitis. Subsequent multimodality imaging demonstrated a right atrial mass arising from the crista terminalis, which was confirmed to be a myxoma at pathologic examination. This case highlights an atypical origin and location of a common cardiac tumor incidentally detected at abdominal imaging.

Purpose To develop and evaluate an interpretable CT spatial phenotype associated with extrapancreatic perineural invasion (EPNI) in resected pancreatic ductal adenocarcinoma. Materials and Methods This multicenter study included training, internal testing, neoadjuvant testing, retrospective external testing, prospective external testing, and The Cancer Imaging Archive (TCIA) sets. The Preoperative Risk of Perineural High-risk Extension in Tumors (PROPHET) model and threshold were locked before testing. Performance was assessed with area under the receiver operating characteristic curve (AUC) and threshold metrics. Six radiologists compared unaided and assisted readings. A stratified Cox model related PROPHET risk to survival, adjusting for age, sex, tumor size, carbohydrate antigen 19-9, nodal category, and differentiation. Results Among 1854 patients (median age, 63 years; 1131 male), AUCs were 0.92 (95% CI, 0.86-0.96) in prospective external testing, 0.85 (95% CI, 0.75-0.93) in retrospective external testing, 0.90 (95% CI, 0.83-0.95) after neoadjuvant therapy, 0.99 (95% CI, 0.98-> 0.99) in internal testing, and 0.80 (95% CI, 0.68-0.93) in TCIA. TCIA sensitivity and specificity were 24/30 (80%) and 2/19 (11%). Assistance increased mean sensitivity to 93% from 82% and reduced reading time to 3.02 from 3.64 minutes per patient (both Holm-adjusted P < .001). Higher PROPHET risk was associated with shorter survival (adjusted hazard ratio, 1.95; 95% CI, 1.53-2.47; nominal P < .001). Conclusion PROPHET provides a CT-visible tumor-vessel spatial phenotype associated with pathologically confirmed EPNI. © The Author(s) 2026. Published by the Radiological Society of North America under a CC BY 4.0 license.

Radiology is adopting artificial intelligence (AI) faster than it is studying what AI does to radiologists' perceptual expertise and interpretive skill. The question has partly shifted from whether these tools perform well on benchmark tasks to what happens to clinical judgment when use of these tools becomes routine. Two risks accompany this transition: automation bias, the tendency to follow AI suggestions without adequate scrutiny, and deskilling, the erosion of perceptual and interpretive ability through disuse, typically revealed when AI support is withdrawn. Incorrect AI output degrades accuracy regardless of experience, the benefit of assistance varies unpredictably across readers and tasks, and performance falls after AI withdrawal; neither risk has been studied longitudinally in radiology. A confidence-stratified triage model is proposed in which calibrated AI confidence, not findings alone, routes cases to low-, intermediate-, or high-confidence streams: output withheld until independent reading, shown with uncertainty signals and a checklist, or shown with random blinded audit. Demonstrated local confidence calibration is a precondition for deployment, and stream thresholds and audit rates should be weighted by task and potential clinical harm, not error probability alone. Rather than applying one workflow to every case, the model redistributes radiologist attention to where AI error is most likely and most consequential, operationalizing four principles: keeping the radiologist in the decision loop, making AI uncertainty visible, inserting targeted cognitive friction, and governing the human-AI interaction as a system. A research agenda to determine whether these countermeasures reduce automation bias and skill erosion in practice is also proposed.