Radiology最新文献

Acute chest pain poses a substantial burden on health care systems and is a common reason for emergency department visits worldwide. Although diagnostic pathways are available to triage patients-distinguishing those needing immediate intervention from those safe for discharge-accurately diagnosing acute coronary syndrome (ACS) remains challenging. ACS is present in only 5%-10% of acute chest pain cases; however, most patients undergo extensive investigations to rule it out due to the clinical and legal consequences of misdiagnosis. In response, coronary CT angiography (CCTA) has emerged as a promising diagnostic tool, gaining support from international societies over the past decade. The 2023 European Society of Cardiology guidelines recommend considering CCTA or functional imaging tests for the initial evaluation of patients without electrocardiographic changes and with uncertain high-sensitivity cardiac troponin levels (class IIa, level A). Similarly, U.S. guidelines support the use of CCTA or functional tests to exclude clinically significant coronary artery disease in patients with intermediate risk of ACS after inconclusive evaluations (class I, level A). This article examines the role of CCTA in managing different risk populations; the potential for CCTA to optimize resource utilization while maintaining diagnostic accuracy; and the impact of other diagnostic resources, such as high-sensitivity troponins, on established diagnostic pathways.

Background Performance of artificial intelligence (AI)-based fracture detection is often evaluated on test sets that may not reflect real-world practice and overestimate AI accuracy, limiting clinical usability. Purpose To evaluate the impact of test set composition on the performance of AI models used for automated pediatric fracture detection at radiography. Materials and Methods This retrospective study analyzed pediatric appendicular trauma radiographs. Training and validation sets used consecutive radiographs, while two internal test sets were created: a "difficult" set containing radiographs with discrepancies between initial and final assessments and a "matched" set selected based on age, body region, and fracture presence. Three pediatric radiologists independently rated images for diagnostic difficulty and image quality. Sets were evaluated by EfficientNet (classification) and You Only Look Once, version 8 (YOLOv8) (object detection), models trained and validated on radiographs. Statistical significance was assessed with generalized estimating equations and odds ratios (ORs). Results A total of 39 703 pediatric patients were included (median age, 10.4 years [IQR, 6.3-13.3 years]; 56% [22 079 of 39 703] male). The difficult internal test set (392 patients, 786 radiographs) was associated with a 40% decrease in odds of a correct classification compared with the matched internal test set (526 patients, 786 radiographs) for EfficientNet variants (OR, 0.60 [95% CI: 0.53, 0.68]; P < .001). For YOLOv8 variants, difficult images were associated with an 80% decrease in odds of successful detection (OR, 0.20 [95% CI: 0.17, 0.25]; P < .001). Sensitivity analysis confirmed that images from the difficult set were rated as more challenging by experts than the "matched" set, and the difficult set contained more "complex" images (15% [119 of 786] vs 5% [41 of 786]) and fewer "simple" images (49% [381 of 786] vs 70% [550 of 786]) (P < .001). Conclusion AI performance in pediatric fracture detection was influenced by test set composition and radiograph complexity, where an internal test set of complex radiographs was associated with decreased odds of correct prediction. © RSNA, 2026 Supplemental material is available for this article.

Background Restenosis is a major cause of stroke recurrence after bare-metal stent (BMS) placement in patients with symptomatic intracranial atherosclerotic disease (ICAD). Observational studies have shown that drug-coated balloon (DCB) angioplasty can reduce restenosis rates. Purpose To compare the efficacy and safety of DCB angioplasty with that of BMS placement in individuals with symptomatic ICAD with high-grade stenosis. Materials and Methods Eligible patients with symptomatic ICAD at 14 Chinese tertiary hospitals were prospectively and randomly assigned (1:1 ratio) to the DCB and BMS groups. The primary outcome was 6-month restenosis assessed with digital subtraction angiography. Secondary and safety outcomes included 6-month symptomatic restenosis, 30-day to 1-year recurrent ischemic event, and 30-day stroke or death. Between-group differences in outcomes were tested using generalized linear and Cox regression models. Results Between July 2021 and March 2023, 209 participants (median age, 59 years [IQR, 52-66 years]; 157 men), 103 and 106 in the DCB and BMS groups, respectively, were included in the intention-to-treat analysis. A total of 164 participants completed the 6-month digital subtraction angiography follow-up, and 203 participants completed the 1-year clinical follow-up. The 6-month angiographic restenosis rate was lower in the DCB group than in the BMS group (11% vs 29%; risk ratio, 0.38 [95% CI: 0.19, 0.78]; P = .006). The DCB group also had a lower 6-month symptomatic restenosis rate (1% vs 10%; risk ratio, 0.13 [95% CI: 0.02, 0.96]; P = .02) and lower 30-day to 1-year recurrent ischemic event rate (4% vs 13%; hazard ratio, 0.31 [95% CI: 0.10, 0.94]; P = .04). The 30-day stroke or death rate was similar in the DCB and BMS groups (6% vs 5%; hazard ratio, 1.24 [95% CI: 0.38, 4.05]; P = .73). Conclusion In individuals with symptomatic ICAD with high-grade stenosis, DCB angioplasty reduced the 6-month risks of angiographic restenosis and symptomatic restenosis and 30-day to 1-year recurrent ischemic event rate compared with BMS placement. Chinese Clinical Trial Registry no. ChiCTR2100046829 © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Wojak in this issue.

Background Reducing the radiation dose without compromising image quality or diagnostic accuracy is essential for repeat CT monitoring in emphysema. Purpose To evaluate the accuracy of ultra-low-dose (ULD) photon-counting CT (PCCT) for both visually and quantitatively assessing emphysema, and to compare this accuracy with that of low-dose (LD) PCCT. Materials and Methods Participants with emphysema who underwent same-day LD and ULD PCCT between November 2024 and February 2025 were prospectively included. Two radiologists independently evaluated the image quality parameters (overall image quality, sharpness, artifacts, and noise) and visually assessed emphysema (subtype and severity). Automated emphysema quantification was performed using low-attenuation volume (LAV) analysis. For centrilobular emphysema (CLE), automated LAV measurements were converted to severity grades and compared with visual severity grades. Paired t tests, Cohen κ analysis, and intraclass correlation coefficients were used to evaluate differences and agreement in findings at LD and ULD PCCT. Results In 152 participants (median age, 68 years [IQR, 60-72 years]; 139 male participants), the ULD protocol reduced radiation exposure by 87% compared with the LD protocol (mean effective dose, 0.20 mSv ± 0.03 [SD] vs 1.58 mSv ± 0.39; P < .001). There was no evidence of a difference between the two protocols in overall image quality (median score, 4 [IQR, 4-5] for both; P = .16), sharpness (median score, 4 [IQR, 4-5] for both; P = .08), or artifacts (median score, 4 [IQR, 3-4] for both; P = .39), but ULD images had lower noise scores (ie, more noise) than LD images (median score, 3 [IQR, 3-3] vs 4 [IQR, 4-4]; P < .001). There was excellent agreement between the two protocols for grading of visual CLE severity (weighted κ = 0.98) and paraseptal emphysema severity (κ = 0.96). The two protocols exhibited excellent agreement in LAV measurements across the lungs and for individual lung lobes (intraclass correlation coefficient range, 0.96-0.98). Visual CLE severity grades demonstrated good agreement with LAV measurements for both the LD (weighted κ = 0.73) and ULD (weighted κ = 0.75) protocols. Conclusion PCCT enables accurate visual and automated assessments of emphysema at a radiation dose equivalent to that of two chest radiographs without compromising image quality. © RSNA, 2026 Supplemental material is available for this article.

Background Previous clinical trials have supported the use of endovascular therapy (EVT) for basilar artery occlusion (BAO) stroke within 24 hours of symptom onset. However, the safety and effectiveness of EVT in patients with BAO treated beyond 24 hours remains unclear. Purpose To compare clinical outcomes and safety following EVT combined with standard medical care versus medical care alone in patients with acute ischemic stroke due to BAO treated beyond 24 hours from symptom onset. Materials and Methods This multicenter retrospective study enrolled patients between March 2017 and April 2024 across China. Eligible patients had BAO and were treated with EVT or standard medical care beyond 24 hours from symptom onset. The primary outcome was the proportion of patients achieving good functional status (modified Rankin Scale score, 0-3). Primary safety outcomes included 90-day mortality and symptomatic intracranial hemorrhage. Inverse probability-weighted regression was performed to adjust for prespecified clinical characteristics, and instrumental variable analysis was repeated as sensitivity analysis. Results Among 217 patients (median age, 66 years [IQR, 58-73 years]; 160 men), good functional status at 90 days was achieved in 35.7% (46 of 129) of patients who underwent EVT and 26.1% (23 of 88) of controls (inverse probability of treatment weighting [IPTW]-adjusted risk ratio [RR], 1.67; P = .008), which was confirmed by instrument variable analysis (adjusted RR, 2.18; P = .04). There was no evidence of a difference in mortality at 90 days between EVT and control groups (48.1% [62 of 129 patients] vs 54.6% [48 of 88 patients]; IPTW-adjusted RR, 0.80; P = .10). Risk of symptomatic intracranial hemorrhage was higher in EVT compared with control groups (11.9% [15 of 126 patients] vs 1.3% [one of 80 patients]; IPTW-adjusted RR, 11.01; P = .02). Conclusion In this study of patients with BAO treated beyond 24 hours from symptom onset, EVT was associated with higher odds of good functional status at 90 days compared with standard medical care, albeit with increased odds of symptomatic intracranial hemorrhage. Chinese Clinical Trial Registry no. ChiCTR2000041117 © RSNA, 2026 Supplemental material is available for this article.