Background Ordered-subset expectation-maximization (OSEM) PET reconstruction is limited by spatial resolution and partial-volume effects, contributing to interreader variability and false positives and highlighting the need for more comprehensive alternative approaches. Purpose To evaluate whether MRI-guided brain PET reconstruction improves inter-reader agreement, visual consistency, diagnostic performance, and Centiloid (CL)-based quantitative separation of amyloid status compared with OSEM reconstruction. Materials and Methods In this retrospective multireader diagnostic accuracy study conducted at a single institution, consecutive brain fluorine 18 (18F)-florbetapir PET/MRI scans (September 2019 to December 2024) were analyzed. For each scan, OSEM and MRI-guided PET reconstructions were generated. Four blinded readers independently interpreted both reconstructions in separate sessions, recording amyloid status (β-amyloid [Aβ]-/Aβ+), diagnostic confidence (three-point scale), and image quality (five-point scale). The consensus-derived reference standard (Aβ-/Aβ+) was not fully independent. Diagnostic performance was assessed using generalized linear mixed-effects models. Interreader agreement was assessed using Fleiss κ, and quantitative measures (CL values and standardized uptake value ratios) were compared between reconstructions. P values were adjusted for multiple comparisons. Results A total of 120 patients (120 scans; median age, 74.8 years [IQR, 68.7-78.3 years]; 66 males) were included. MRI-guided reconstruction improved interreader agreement for amyloid status (Fleiss κ, 0.91 vs 0.79; absolute increase, 0.12 [95% CI: 0.05, 0.20]; P < .001), with greatest improvement in the posterior cingulate cortex and precuneus (κ, 0.82 vs 0.74; P = .03). Specificity improved from 91.6% (95% CI: 85.7, 96.5) to 99.2% (95% CI: 98.4, 99.7), reducing false positives from 8.1% (14 of 166) to 0.8% (two of 168) (P < .001) relative to the reference standard. CL separation was preserved between Aβ- and Aβ+ groups (MRI-guided: -4.24 vs 89.74; OSEM: 0.84 vs 89.0; both P < .001), with lower median CL values in Aβ- scans (-4.24 vs 0.84; P = .01). Conclusion MRI-guided amyloid brain PET reconstruction showed higher interreader agreement and fewer false-positive classifications than OSEM reconstruction and preserved CL-based separation between the Aβ- and Aβ+ groups. © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Bedmutha in this issue.
Background Radiographs sometimes do not depict femoral-neck fractures, particularly radiograph-negative or indeterminate femoral-neck fractures, leading to delayed treatment and complications. Purpose To develop and externally evaluate a deep learning model, OccuNet, for detecting femoral-neck fractures on pelvic or hip radiographs, compare its performance and accuracy with those of radiologists and emergency medicine physicians, and evaluate its effect on reader performance and accuracy. Materials and Methods This multicenter retrospective study included adults suspected of having hip trauma who underwent pelvic or hip radiography and same-episode CT or MRI at four hospitals (January 2009-August 2025). Patients were split into a training set, an internal test set, and three external test sets. A two-stage model, OccuNet, was trained: Stage 1 used contrastive pretraining on paired original and artifact-augmented versions of the same radiograph to learn artifact-robust features, and stage 2 fine-tuned a fracture detector. Performance and accuracy were compared with those of 10 readers (five musculoskeletal radiologists, five emergency medicine physicians), and artificial intelligence (AI) assistance was tested for sensitivity, specificity, area under the receiver operating characteristic curve (AUC), and reading time. Results This study included 2576 patients (mean age [±SD], 69 years ± 11.2; 1380 women). In the pooled group (n = 1766), the model achieved a sensitivity of 97.5% (913 of 936), specificity of 98.8% (820 of 830), and excellent performance (AUC, 0.99 [95% CI: 0.99, 0.99]). For radiograph-negative or indeterminate fractures (n = 189), sensitivity of the model (94.7% [179 of 189]) was higher than that of radiologists (86.2% [163 of 189]; P < .001) and emergency medicine physicians (68.8% [130 of 189]; P < .001). With AI assistance, radiologist and emergency medicine physician sensitivities increased from 93.7% (877 of 936) to 97.2% (910 of 936; P < .001) and from 84.3% (789 of 936) to 95.6% (895 of 936; P < .001), respectively, and mean reading times shortened by 14.9% and 18.9% (both P < .001), respectively. Conclusion OccuNet demonstrated excellent performance for femoral-neck fracture detection on pelvic or hip radiographs, maintained high sensitivity in radiograph-negative or indeterminate (Garden I-II) fractures, and improved reader sensitivity and efficiency. © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Flores and Cantarelli in this issue.
Background Opioid-benzodiazepine regimens remain common for radiologist-administered procedural sedation despite respiratory and analgesic effectiveness concerns. Purpose To compare intraprocedural pain and patient-reported experience between ketamine/midazolam and fentanyl/midazolam during image-guided procedural sedation. Materials and Methods This randomized clinical trial was conducted at a single academic center between June 2025 and February 2026. Adults undergoing image-guided lung or bone biopsy or abscess drainage were randomized to fentanyl/midazolam or ketamine/midazolam administered by interventional radiologists. Procedures were performed using US, CT, CT fluoroscopy, or combined CT and US guidance. The primary outcome was maximum intraprocedural pain (0-10 on the Numeric Rating Scale). Secondary outcomes included sedation depth, physiologic parameters, oxygen desaturation, patient-reported experience assessed using a modified Heidelberg questionnaire, and complications. Results Among 264 randomized procedures (132 procedures per group) in 260 participants (median age, 68 years [IQR, 61-75 years]; 135 [52%] female), ketamine/midazolam resulted in lower maximum intraprocedural pain than fentanyl/midazolam (mean difference, -1.4 points [95% CI: -2.0, -0.8]; P < .001). Pain scores greater than 4 occurred less frequently with ketamine/midazolam (2.3% vs 17%; absolute difference, 14 percentage points [95% CI: 8, 21]; P < .001). Ketamine/midazolam was associated with higher nadir oxygen saturation (mean difference, +1.4% [95% CI: 0.6, 2.2]; P = .001) and fewer oxygen desaturation events below 90% (three [2.3%] vs 13 [9.8%]; absolute difference, 7.6 percentage points [95% CI: 1.9, 13.3]; P = .02). Ketamine/midazolam produced deeper sedation and higher intraprocedural systolic blood pressure. Hallucinations occurred more frequently with ketamine/midazolam (15 [11.4%] vs five [3.8%]; absolute difference, 7.6 percentage points [95% CI: 1.3, 13.9]; P = .03), though overall procedural comfort, reduced recall, and perceived adequacy of sedation were improved. Procedure-related and sedation-related complications did not differ between groups. Conclusion Radiologist-administered ketamine/midazolam during image-guided procedural sedation improved analgesia and patient-reported experience with fewer hypoxemic events and no increase in complications compared with fentanyl/midazolam. Clinical trial registration no. NCT07040163 © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Weiss and Park in this issue.
Background Measurement of Crohn disease (CD) activity is crucial for guiding treatment decisions. Super-resolution contrast-enhanced US (SRCEUS) is an emerging technique that can be used to assess vascularity at micrometer scale; however, its performance in differentiating CD activity stages is not yet known. Purpose To evaluate SRCEUS in the assessment of disease activity in participants with CD. Materials and Methods This prospective study included participants with CD involving the terminal ileum or colon who underwent intestinal SRCEUS imaging between January 2025 and June 2025. Participants were classified into remission/mild or moderate/severe activity groups based on the Simple Endoscopic Score for Crohn's Disease. B-mode US, color Doppler flow imaging (CDFI), conventional contrast-enhanced US (CEUS), and SRCEUS were performed. SRCEUS parameters were calculated. Logistic regression modeling, receiver operating characteristic analysis, and the DeLong test were used to evaluate and compare diagnostic performance. Results This study included 54 participants (mean age, 30 years ± 8.5 [SD]; 42 men). Key SRCEUS vascular parameters were higher in the moderate/severe activity group than the remission/mild activity group, including maximum density (mean, 63.4 ± 13.1 vs 51.1 ± 15.6; P = .003), mean density (mean, 26.9 ± 5.3 vs 21.1 ± 8.0; P = .007), mean velocity (mean, 40.2 mm/sec ± 27.3 vs 23.0 mm/sec ± 14.9; P = .01), vascular density ratio (mean, 73.2% ± 12.8 vs 38.5% ± 17.8; P < .001), and fractal dimension (mean, 1.7 ± 0.1 vs 1.5 ± 0.1; P < .001). SRCEUS performed well in distinguishing participants with remission/mild activity from those with moderate/severe activity (area under the receiver operating characteristic curve [AUC], 0.92; sensitivity, 94.1%; specificity, 95.0%). SRCEUS performed better than conventional CEUS (AUC, 0.78; P = .03). Clinical, B-mode US, CEUS, and SRCEUS parameters were included in a combined model, in which the International Bowel Ultrasound Segmental Activity Score and vascular density ratio were identified as influential factors. The combined model had an AUC of 0.94, demonstrating superior performance to both CDFI (AUC, 0.82; P = .04) and conventional CEUS (P = .002) and similar performance to SRCEUS (P = .51). Conclusion SRCEUS demonstrated excellent performance in differentiating CD activity through high-resolution depiction of microstructural changes, with superior performance to conventional CEUS. © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Katwal and Chernyak in this issue.
Hepatocellular carcinoma (HCC) remains a leading cause of cancer mortality, with limited curative options. Histotripsy is a noninvasive focused ultrasound ablation technique that destroys tumors through acoustic cavitation rather than heat-based methods, offering an alternative to radiofrequency and microwave ablation. This review traces histotripsy from preclinical development to U.S. Food and Drug Administration (FDA) approval for liver tumors. Unlike thermal methods, histotripsy creates precise mechanical tissue fractionation while preserving critical structures, including blood vessels and bile ducts. The technology uses short, high-pressure ultrasound pulses to generate cavitation bubble clouds that liquefy targeted tissue without thermal damage. Preclinical studies demonstrated complete tumor ablation with robust immune cell infiltration, suggesting potential abscopal effects. Clinical trials, including THERESA and #HOPE4LIVER, confirmed safety, with #HOPE4LIVER demonstrating short-term efficacy, achieving approximately 95% technical success and approximately 90% 1-year local control rates. Histotripsy received FDA approval in October 2023 for treating liver cancer. The technology shows promise for perivascular lesions where heat-sink effects compromise thermal ablation. Current evidence supports optimal outcomes for tumors 3 cm or smaller, with ongoing research exploring combinations with immunotherapy and applications beyond HCC. Histotripsy represents a paradigm shift in local ablation, offering mechanical precision combined with potential systemic immunostimulatory benefits. © The Author(s) 2026. Published by the Radiological Society of North America under a CC BY 4.0 license.
Background Photon-counting CT (PCCT) offers improved spatial resolution, contrast to noise ratio, and dose efficiency, but its clinical utility remains incompletely defined for breast cancer. Purpose To evaluate the feasibility of PCCT for pretreatment breast cancer assessment through comparisons with MRI, full-field digital mammography (FFDM), and fluorine 18 (18F) fluorodeoxyglucose (FDG) PET/CT. Materials and Methods In this prospective study (March-May 2025), female participants with breast lesions categorized as Breast Imaging Reporting and Data System 4C or higher at US or FFDM underwent breast MRI and multiphasic contrast-enhanced PCCT. 18F-FDG PET/CT was performed in a subset with locally advanced disease. Four radiologists independently evaluated lesion morphologic characteristics, additional findings, and clinical TNM stage. Agreement was analyzed using intraclass correlation coefficients (ICCs) and κ statistics. The diagnostic performance for additional lesions and nodal metastasis was compared with the reference standard (pathologic examination). Results Among 126 participants (mean age, 58.1 years ± 12.3 [SD]), interreader agreement across PCCT, MRI, and FFDM was good to excellent. PCCT agreed with MRI for lesion characterization (κ = 0.57-0.96) and clinical T categorization (κ = 0.86-0.88), with highest agreement with pathologic size (ICC, 0.70-0.81). For 46 pathologically confirmed additional lesions, PCCT was more sensitive than FFDM (difference, 44% [95% CI: 19, 66]) and similar to MRI (difference, 7% [95% CI: -5, 21]). Additionally, 44% (95% CI: 27, 52) of microcalcifications were missed at PCCT versus FFDM. For pathologically confirmed nodal metastasis, PCCT was more sensitive (difference, 10% [95% CI: 1, 20]) and accurate (difference, 6% [95% CI: 1, 11]) than MRI. For clinical N category, PCCT agreed with PET/CT (κ = 0.82 [95% CI: 0.62, 0.96]; n = 19). Two distant metastases identified at PCCT were consistent with 18F-FDG PET/CT and pathologic findings. Conclusion PCCT demonstrated similar performance to MRI for lesion characterization and detection of additional lesions, with better performance for nodal metastasis evaluation; however, detection of microcalcifications was limited. © RSNA, 2026 Supplemental material is available for this article.
Background Chest CT is a primary method for identifying pulmonary nodules, yet interpreting scans remains time-intensive and demanding. Currently, artificial intelligence (AI) is expected to reduce reading times, but the effect of AI on reporting times in this setting is unknown. Purpose To evaluate the impact of a commercial AI software on radiologists' reading time for pulmonary nodule assessment on chest CT scans within a real-world clinical setting. Materials and Methods This retrospective study included patients who underwent chest CT examinations at a tertiary medical center between September 2021 and May 2024. The study period was divided into pre- and post-AI phases. The primary outcome was radiology reporting time. The association between AI implementation and reporting time was evaluated using a multivariable parametric Weibull shared frailty survival model adjusted for reader function, examination type, patient location, and requesting specialty, with clustering at the radiologist level. Interaction analyses assessed heterogeneity across prespecified subgroups. An exploratory extrapolation estimated projected workforce and financial impact. Results This study included 19 433 patients (mean age, 62 years ± 14.2 [SD]; 21 814 men; 39 323 chest CT examinations, 19 190 pre-AI, and 20 133 post-AI). AI implementation was associated with faster report completion (adjusted hazard ratio, 1.17; 95% CI: 1.14, 1.21; P < .001). The adjusted median reporting time decreased from 21.3 minutes pre-AI to 18.2 minutes post-AI (14.6% reduction; P < .001). Heterogeneity was observed across reader function (P < .001), examination type (P = .048), and requesting specialty (P = .03). The largest relative reductions were observed for CT thorax electrocardiogram-gated examinations (-41.1%; P < .001) and thoracic radiologists (-25.0%; P < .001), whereas emergency department examinations showed increased median reporting time (7.1%; P < .001). At institutional scan volumes (approximately 20 000-22 000 chest CT examinations annually), exploratory modeling suggested an approximate reduction of 0.5 full-time equivalent radiologist workload. Conclusion Implementation of commercial AI-assisted pulmonary nodule assessment on chest CT scans reduced radiologist reporting time in a real-world clinical setting. © 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 Iwasawa in this issue.
Background Ferumoxytol has been described as an alternative contrast agent for vascular suppression in MR neurography (MRN), but its diagnostic utility in patients has yet to be evaluated. Purpose To evaluate the impact of ferumoxytol on vascular suppression, nerve conspicuity, and evaluation of nerve abnormalities at three-dimensional (3D) brachial plexus MRN, compared with noncontrast and gadolinium-enhanced MRN, in participants with suspected Parsonage-Turner syndrome (PTS) or thoracic outlet syndrome (TOS). Materials and Methods This prospective study included participants who underwent 3D MRN with and/or without gadolinium chelate for clinical suspicion of PTS or TOS and subsequently underwent 3D MRN with ferumoxytol (within 3 months of the clinical examination). Two musculoskeletal radiologists qualitatively evaluated 3D short-tau inversion-recovery fast spin-echo scans for the degree of vascular suppression, nerve conspicuity, and presence of nerve abnormalities. Wilcoxon signed-rank or McNemar tests were used for comparing noncontrast and gadolinium-enhanced scans with ferumoxytol-enhanced scans. Results This study included 18 participants (mean age, 42 years ± 15.2 [SD]; 10 men). Ferumoxytol-enhanced scans demonstrated improved vascular suppression compared with both noncontrast scans (both raters, P < .001) and gadolinium-enhanced scans (both P = .04). For rater 2, ferumoxytol-enhanced acquisitions demonstrated improved conspicuity of several nerve segments relative to the noncontrast scan, including segments of the suprascapular (P = .01), axillary (P = .02), and long thoracic nerves (P = .004). The distribution of scores for these nerve segments for rater 1 also favored ferumoxytol-enhanced versus noncontrast scans, but the differences were not statistically significant (all P ≥ .06). There was no evidence of a difference in nerve conspicuity between ferumoxytol-enhanced and gadolinium-enhanced scans (P ≥ .17 for all nerve segments) and also no evidence of discrepancies in abnormal nerve findings between the acquisitions (all P ≥ .48). Conclusion Ferumoxytol improved vascular suppression compared with noncontrast and gadolinium-enhanced 3D short-tau inversion-recovery fast spin-echo sequences in brachial plexus MRN, enhancing the conspicuity of several nerve branches versus noncontrast scans, with similar detection of abnormal nerve findings. © RSNA, 2026 Supplemental material is available for this article.
